1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
|
ArrowB.c Current rev 1.24 Last Release 1.22
ArrowB.c 1.24
date: 1998/10/10 07:40:44; author: gritton; state: Exp; lines: +6 -6
Fix _XmDrawArrow for proper shadow thickness on 2.0.
ArrowButton: shadow should always be 2.
ScrollBar: shadow should be 0, 1, or 2.
ArrowB.c 1.23
date: 1998/09/21 22:10:36; author: gritton; state: Exp; lines: +4 -4
The ArrowButton (and gadget) comes up with the wrong initial size.
Unlike the other widgets, the size is absolute, not taking shadow and
highlight thicknesses into account.
ArrowBG.c Current rev 1.21 Last Release 1.19
ArrowBG.c 1.21
date: 1998/10/10 07:40:44; author: gritton; state: Exp; lines: +6 -6
Fix _XmDrawArrow for proper shadow thickness on 2.0.
ArrowButton: shadow should always be 2.
ScrollBar: shadow should be 0, 1, or 2.
ArrowBG.c 1.20
date: 1998/09/21 22:10:37; author: gritton; state: Exp; lines: +4 -7
The ArrowButton (and gadget) comes up with the wrong initial size.
Unlike the other widgets, the size is absolute, not taking shadow and
highlight thicknesses into account.
BulletinBoard.c Current rev 1.38 Last Release 1.32
BulletinBoard.c 1.38
date: 1998/10/23 17:27:39; author: rwscott; state: Exp; lines: +4 -3
BulletinBoard.c:change_managed() - Always do the layout, whether we are
realized or not.
MainW.c:realize() - Get rid of the geo stuff in here. It gets done in
change_managed, which is call just before realize.
ScrolledW.c:realize() - Just a comment for now. I don't think this geometry
stuff should be happening here.
This is all to get mainw/test13 working. It's not correct yet, the sizes are
still off but, at least it is put together correctly now.
BulletinBoard.c 1.37
date: 1998/10/21 22:33:27; author: rwscott; state: Exp; lines: +11 -3
realize() - We were doing a lot of geometry calculations in here which
should not be necessary. The geo stuff will be done in change_managed,
which is called just before realize.
This fixes up the new bulletinboard/test14
BulletinBoard.c 1.36
date: 1998/10/16 23:41:36; author: rwscott; state: Exp; lines: +5 -3
DebugUtil.c - debug stuff (no kidding)
BulletinBoard.c:query_geometry() - debug stuff
MainW.c:_XmMainWindowPreferredSize() - query the preferred size of the message
window (mainw/test2, 3, 4, 5, 6)
GeoUtils.c:_XmHandleQueryGeometry() - If policy is RESIZE_ANY we can't just
return the intended, we may have started this query off with a NULL
request to find the preferred size in the absence of any constraints.
This makes this look more like the query_geometry of Form.
(mainw/test2, 3, 4, 5, 6, messagebox/test3)
BulletinBoard.c 1.35
date: 1998/09/30 00:15:20; author: rwscott; state: Exp; lines: +8 -2
realize() - bulletinboard/test11 shows that a 0x0 bulletinboard gets realized
to 1x1.
BulletinBoard.c 1.34
date: 1998/09/29 19:55:52; author: jon; state: Exp; lines: +10 -2
Plug memory leak in form
BulletinBoard.c 1.33
date: 1998/09/20 02:07:02; author: rwscott; state: Exp; lines: +7 -2
_XmBbButton() - get rid of the popdownCallback call. This is a job for Shell
not BB.
Jon's bulletinboard/test13 double callback problem.
CascadeB.c Current rev 1.96 Last Release 1.87
CascadeB.c 1.96
date: 1998/10/17 23:09:57; author: rwscott; state: Exp; lines: +25 -6
Pay attention to Lab_RecomputeSize. (cascadebutton/test8)
CascadeB.c 1.95
date: 1998/10/02 14:57:44; author: rwscott; state: Exp; lines: +12 -4
Screen.c - add some instrumentation to Alloc/Free pixmap
CascadeB[G] - make them use the pixmap cache for both allocating and freeing
the arrow pixmaps. Make sure the pixmap instance variables get
initialized.
CascadeB.c 1.94
date: 1998/09/25 03:22:22; author: rwscott; state: Exp; lines: +97 -194
The recent geometry changes in these were killing many other tests. A change
in Label does not just change Label! It changes PushB ToggleB and possibly
some things you don't even realize. For example the tests in just about every
directory use a label or labelgadget or one of its sub-classes. If you make
a change to something like this you had better have some damn good proof that
it is correct _and_ that it does not screw up three others that have been
working. Labels have been around for a _long_ time and have been picked apart
at the pixel level. Because of that they are used as the base for just about
every test in form. All of the form tests are also a good test of label since
they print out the geo of every widget. You make a change to something this
basic and you better be willing to back it up. I could not get through half
of my 30 or so test apps without something being messed up on the initial
screen, and most of these have been working for a _very_ long time.
CascadeB.c 1.93
date: 1998/09/24 20:12:37; author: kbwj; state: Exp; lines: +4 -3
Fixed placement of cascade
CascadeB.c 1.92
date: 1998/09/24 18:36:43; author: kbwj; state: Exp; lines: +4 -4
Fixed some typos with CBG instead of CB. Insert a check if CB_Submenu
is NULL in DrawCascade.
CascadeB.c 1.91
date: 1998/09/24 01:49:27; author: rwscott; state: Exp; lines: +11 -12
Ther was a whole bunch of CBG_ macros being used in here without the
required XmIsGadget.
Karsten, you have to be real carefull with these Gadget things. The CB_ and
CBG_ macros are not the same thing. Same applies to Lab_ and LabG_ and all
the rest.
This gets cascadebutton/test6 going again and, hopefully gets rid of all the
segfaults in ml now.
CascadeB.c 1.90
date: 1998/09/23 04:23:49; author: kbwj; state: Exp; lines: +186 -89
Changed cascade geometry when in an option menu
CascadeB.c 1.89
date: 1998/09/22 18:15:05; author: jon; state: Exp; lines: +87 -184
Backing out a set of recent changes to restore some sanity till we kin
figger out what went wrong.
(The Texans are getting to me!)
CascadeB.c 1.88
date: 1998/09/22 13:27:14; author: kbwj; state: Exp; lines: +186 -89
Changed the drawing and geometry of cascade when in a option menu
CascadeBG.c Current rev 1.85 Last Release 1.78
CascadeBG.c 1.85
date: 1998/10/02 14:57:45; author: rwscott; state: Exp; lines: +21 -4
Screen.c - add some instrumentation to Alloc/Free pixmap
CascadeB[G] - make them use the pixmap cache for both allocating and freeing
the arrow pixmaps. Make sure the pixmap instance variables get
initialized.
CascadeBG.c 1.84
date: 1998/09/25 03:22:22; author: rwscott; state: Exp; lines: +132 -215
The recent geometry changes in these were killing many other tests. A change
in Label does not just change Label! It changes PushB ToggleB and possibly
some things you don't even realize. For example the tests in just about every
directory use a label or labelgadget or one of its sub-classes. If you make
a change to something like this you had better have some damn good proof that
it is correct _and_ that it does not screw up three others that have been
working. Labels have been around for a _long_ time and have been picked apart
at the pixel level. Because of that they are used as the base for just about
every test in form. All of the form tests are also a good test of label since
they print out the geo of every widget. You make a change to something this
basic and you better be willing to back it up. I could not get through half
of my 30 or so test apps without something being messed up on the initial
screen, and most of these have been working for a _very_ long time.
CascadeBG.c 1.83
date: 1998/09/24 20:12:38; author: kbwj; state: Exp; lines: +4 -3
Fixed placement of cascade
CascadeBG.c 1.82
date: 1998/09/24 18:37:13; author: kbwj; state: Exp; lines: +4 -4
Fixed some typos with CB instead of CBG. Insert a check if CB_Submenu
is NULL in DrawCascade.
CascadeBG.c 1.81
date: 1998/09/23 04:23:48; author: kbwj; state: Exp; lines: +212 -130
Changed cascade geometry when in an option menu
CascadeBG.c 1.80
date: 1998/09/22 18:15:05; author: jon; state: Exp; lines: +128 -210
Backing out a set of recent changes to restore some sanity till we kin
figger out what went wrong.
(The Texans are getting to me!)
CascadeBG.c 1.79
date: 1998/09/22 13:27:46; author: kbwj; state: Exp; lines: +212 -130
Changed the drawing and geometry of cascade when in a option menu
Command.c Current rev 1.14 Last Release 1.11
Command.c 1.14
date: 1998/10/23 01:09:49; author: rwscott; state: Exp; lines: +6 -2
XmCreateCommandDialog() - test2 was showing the wrong widgets. We were
creating the dialogShell with CreateWidget instead of
XmCreateDialogShell, which puts the stuff on the PopupList. test2
now has the correct widget tree.
Command.c 1.13
date: 1998/10/02 17:56:30; author: rwscott; state: Exp; lines: +21 -3
Command.c:initialize() - Since Com_PromptString and ComCommand are just
re-named SelectionBox resources, let SelectionBox deal with them.
Command.c:set_values() - ditto
Command.c:destroy() - ditto
SelectionBox.c:initialize() - if the textString resource is specified make sure
it gets stuffed into the TextField
I suspect that we have a few of these kick'n around. Basically if a class
uses one of its superclasses resources with a name change, the super-class
will take care of anything to do with that resource.
Command.c 1.12
date: 1998/09/29 21:26:49; author: jon; state: Exp; lines: +17 -3
Adding destroy method to free some strings. This cuts back, but doesn't
emliminate the leaking in command/test3
DebugUtil.c Current rev 1.39 Last Release 1.35
DebugUtil.c 1.39
date: 1998/10/16 23:41:37; author: rwscott; state: Exp; lines: +6 -2
DebugUtil.c - debug stuff (no kidding)
BulletinBoard.c:query_geometry() - debug stuff
MainW.c:_XmMainWindowPreferredSize() - query the preferred size of the message
window (mainw/test2, 3, 4, 5, 6)
GeoUtils.c:_XmHandleQueryGeometry() - If policy is RESIZE_ANY we can't just
return the intended, we may have started this query off with a NULL
request to find the preferred size in the absence of any constraints.
This makes this look more like the query_geometry of Form.
(mainw/test2, 3, 4, 5, 6, messagebox/test3)
DebugUtil.c 1.38
date: 1998/10/13 00:43:11; author: rwscott; state: Exp; lines: +29 -2
Add some ScrolledW instrumentation
Don't un-manage the scrollbar if it should not be displayed, just push it off
the edge (scrolledwindow/test10).
Minor adjustments to the size and position of the clipWindow.
This gets about 4 more of the scrolledwindow tests passing.
DebugUtil.c 1.37
date: 1998/10/03 14:49:05; author: rwscott; state: Exp; lines: +10 -2
Trying to find out why we have lost nedit, among one of the first apps that
we ever got working.
Add some instrumentation in aid of the above goal. It turns out that it is
because of the massive change to MainW.c rev 1.39. Specifically the resize
call in the set_values method. We can suggest a new size in this method by
changing XtWidth/XtHeight _but_ we cannot force such a change by calling
resize. This not only lost nedit for us but _many_ of the tests in mainw!
Now to go through the diff from 1.38 to 1.39 and figure out what can stay and
what has to go. This may take awhile!
DebugUtil.c 1.36
date: 1998/09/21 17:19:16; author: jon; state: Exp; lines: +14 -8
Avoid slightly expensive getenvs call if DEBUGSOURCES and
DEBUG_PRINT_WIDGETID aren't defined by remembering if we've already
checked for them.
DialogS.c Current rev 1.16 Last Release 1.12
DialogS.c 1.16
date: 1998/10/18 22:00:57; author: rwscott; state: Exp; lines: +7 -4
DialogS.c - realize the shell in the initialize method instead of the
insert_child method.
Form.c - back out yesterdays change, since it was screwing up xephem, and the
real problem was in DialogS.
Manager.c - instrumentation
PanedW.c - instrumentation
This is wrt form/test43, 42 and Xquote.
The problem was that the DialogShell was being realized in its insert_child
method. This triggered a change_managed call to the paned window, which
would configure the form _before_ the changed_managed of form. Thus the
form children where the wrong size.
DialogS.c 1.15
date: 1998/09/22 18:41:51; author: rwscott; state: Exp; lines: +3 -2
Damn, missed a comment character
DialogS.c 1.14
date: 1998/09/22 18:28:18; author: rwscott; state: Exp; lines: +8 -2
Take out the bit to Unmap the window if we get a map notify and the shell
is not marked as popped up. Somehow nedit's File->Open dialog is getting
displayed without Shell_PoppedUp being True, and this is immediately
un-mapping it.
DialogS.c 1.13
date: 1998/09/20 17:40:41; author: rwscott; state: Exp; lines: +6 -2
StructureNotifyHandler() - If we get a MapNotify event but the Shell is not
popped up, this means that we have XtPoppedDown the dialog before all
the window manager re-parenting. In this case XtUnmap the widget
again.
This helps Xlt/tests/WaitTillMapped/test1
DragBS.c Current rev 1.10 Last Release 1.8
DragBS.c 1.10
date: 1998/09/22 22:03:42; author: jon; state: Exp; lines: +10 -3
Further testing with Netscape 4.5 revealed that these functions are
referenced. Again, here they are just stubs. If anyone can fill these
in, feel free.
int _XmDSIAddChild(){return NULL;}
int _XmDSIRemoveChild(){return NULL;}
int _XmDSIDestroy(){return NULL;}
int _XmDSIGetBorderWidth(){return NULL;}
int _XmDSIReplaceChild(){return NULL;}
int _XmDSIGetChildPosition(){return NULL;}
int _XmDSISwapChildren(){return NULL;}
DragBS.c 1.9
date: 1998/09/22 21:03:48; author: jon; state: Exp; lines: +15 -2
A few (ifdef'ed) fixes for testing binary compatibility with the
dynamically linked Netscape 4.5PR1 release.
Add stub for _XmDSIAddChild() which Netscape wants. No idea about args or
anything, I just return NULL, and get by ok.
_XmTargetsToIndex(): For compatibility testing, return from this routine
immediately with 0 exit code. Netscape was dying in this routine because
it was getting passed a huge (presumably invalid) numTargets value and
attempting to allocate a block of memory based on it. I think Netscape is
calling this routine directly, because I tried catching all the places
where LessTif calls this routine and didn't have any luck.
After making these changes, #defining NS45PR1, and compiling, NS 4.5PR1 it
actually works under LessTif. I got a core dump when I tried to call up a
FSB, but I successfully loaded the netscape home page. We must be doing
*something* right!
-Jon
DragDrop.c Current rev 1.7 Last Release 1.6
DragDrop.c 1.7
date: 1998/09/22 02:25:51; author: rwscott; state: Exp; lines: +24 -21
DragDrop.c:XmDragStart() - If the initiatorProtocolStyle is XmDRAG_NONE
return a NULL DragContext widget.
Label.c, LabelG.c, List.c - Deal with a NULL DragContext widget returned from
XmDragStart.
This allows you to turn off Dragging with the
*dragInitiatorProtocolStyle: XmDRAG_NONE
resource.
DragOverS.c Current rev 1.13 Last Release 1.12
DragOverS.c 1.13
date: 1998/10/19 23:56:28; author: jon; state: Exp; lines: +110 -3
adding in zap_cursor contributed by Gilles Seguin.
DrawingArea.c Current rev 1.30 Last Release 1.26
DrawingArea.c 1.30
date: 1998/10/20 19:41:36; author: rwscott; state: Exp; lines: +4 -3
scrolledwindow/test11 was not resizing the drawing area correctly. This is
a scrolledwindow in AUTOMATIC mode. This change gets test11 working, but
needs a little more looking into. I want to pick it up at another location,
so I am committing it now.
DrawingArea.c 1.29
date: 1998/10/17 20:03:37; author: rwscott; state: Exp; lines: +4 -2
DrawingArea.c:change_managed() - I we are not realized and the width/height
is not zero use that instead of the calculated size. This allows
setting the size in a resource. (rowcolumn/test20)
_XmListSetSBManageHoriz() - if the SizePolicy is CONSTANT we were managing the
scrollbar without paying attention to the SBDisplayPolicy.
This now gets the correct widgets managed for many of the mainw, selectionbox
and command tests.
DrawingArea.c 1.28
date: 1998/10/07 22:31:19; author: rwscott; state: Exp; lines: +25 -6
Manager.c:realize() - drawingarea/test7 shows that if the width/height is 0
then we bump it up to 1, to avoid X errors, but we do not set it
back to 0 after the XtCreateWindow.
DrawingArea.c:initialize() - drawingarea/test7 shows that we do not set the
size here if one is not specified.
DrawingArea.c:PreferedSize() - If we have no children, the preferred size
is our current size, if not 0 or the DA_MarginWidth if it is. (test7)
DrawingArea.c 1.27
date: 1998/10/06 18:42:00; author: pgw; state: Exp; lines: +17 -2
Added code to initialize new 2.0 resources only for 2.0 builds.
DrawnB.c Current rev 1.29 Last Release 1.26
DrawnB.c 1.29
date: 1998/10/24 22:28:36; author: rwscott; state: Exp; lines: +34 -2
DrawnB.c:expose() - Calling the Label expose method caused problems so it was
taken out. mfm shows a clipboard pixmap in a drawn button that seems
to need it, so we steal a bit of the label expose method here to
copy the pixmap if there is one.
Form.c:query_geometry() - Use the same logic as change_managed here to set
the forms size for the layout. Use the current dimension if not
realized, or 0 if we are. This way we honor user preferences.
This finally gets mfm displaying correctly.
DrawnB.c 1.28
date: 1998/09/26 21:58:04; author: rwscott; state: Exp; lines: +7 -3
plug a hole shown by drawnbutton/test6.
DrawnB.c 1.27
date: 1998/09/21 16:18:18; author: rwscott; state: Exp; lines: +6 -2
expose() - This was calling the Label expose method with in turn was calling
the resize method which in turn causes an expose which will cause
a resize...(anyone see a problem yet). Take out the call to the
superclass expose method.
resize() - take out the shadow drawing stuff.
FileSB.c Current rev 1.59 Last Release 1.55
FileSB.c 1.59
date: 1998/10/02 04:17:24; author: rwscott; state: Exp; lines: +189 -503
Cleaned out a lot of baggage, rearranged some stuff. I make the hole a little
smaller, but not much. There is something strange going on, I suspect
because XmNfileListItems and XmNlistItems share the same space.
FileSB.c 1.58
date: 1998/09/20 04:29:19; author: rwscott; state: Exp; lines: +10 -3
mmochol@elka.pw.edu.pl
defaultQualifySearchDataProc() - Problems displaying the root directory, ie
/*
FileSB.c 1.57
date: 1998/09/20 00:51:46; author: rwscott; state: Exp; lines: +3 -3
FileSB.c:initialize() - No need to remove all the default action callbacks
now since SelectionBox is not adding one.
SelectionBox.c:_XmSelectionBoxCreateList() - Don't add a defaultAction callback
to the list. This is now handled with ParentProcess.
FileSB.c 1.56
date: 1998/09/20 00:38:49; author: rwscott; state: Exp; lines: +5 -3
List.c:ListDefaultAction() - If the parent of the list is a manager invoke
the parent process method.
List.c:ListKbdActivate() - simply call ListDefaultAction.
FileSB.c:initialize() - don't install the defaultAction callback on the file
list. ParentProcess is taking care of this now.
This gets rid of the double okCallback in the FileSB
Form.c Current rev 1.123 Last Release 1.116
Form.c 1.123
date: 1998/10/24 22:28:37; author: rwscott; state: Exp; lines: +4 -4
DrawnB.c:expose() - Calling the Label expose method caused problems so it was
taken out. mfm shows a clipboard pixmap in a drawn button that seems
to need it, so we steal a bit of the label expose method here to
copy the pixmap if there is one.
Form.c:query_geometry() - Use the same logic as change_managed here to set
the forms size for the layout. Use the current dimension if not
realized, or 0 if we are. This way we honor user preferences.
This finally gets mfm displaying correctly.
Form.c 1.122
date: 1998/10/24 04:35:00; author: rwscott; state: Exp; lines: +7 -4
constraint_set_values() - Okay, it seems that you do have to re-do the geo
in here but, only if a constraint that affects layout changes _and_
the form is realized _and_ the form is managed.
3 of the variations that show this are scrolledwindow/test13, form/test15 and,
Xinvest when it is centering the lower right hand frame. A pattern may just
be starting to emerge here!!!!!
Form.c 1.121
date: 1998/10/23 22:49:13; author: rwscott; state: Exp; lines: +5 -2
constraint_set_values() - Never do the geometry request in here.
(scrolledwindow/test13)
Form.c 1.120
date: 1998/10/18 22:00:58; author: rwscott; state: Exp; lines: +2 -6
DialogS.c - realize the shell in the initialize method instead of the
insert_child method.
Form.c - back out yesterdays change, since it was screwing up xephem, and the
real problem was in DialogS.
Manager.c - instrumentation
PanedW.c - instrumentation
This is wrt form/test43, 42 and Xquote.
The problem was that the DialogShell was being realized in its insert_child
method. This triggered a change_managed call to the paned window, which
would configure the form _before_ the changed_managed of form. Thus the
form children where the wrong size.
Form.c 1.119
date: 1998/10/17 13:28:25; author: rwscott; state: Exp; lines: +6 -2
PanedW.c:realize() - Take out the geo negotiations, this will get done in
change_managed.
Form.c:XmFormFindPreferred() - Do not set the FCP_Pref[W|H] in here! Since
RC is initializing with a size of 16x16 now, Form was taking this
as a preferred size and laying out to this! (form/test42)
This gets the Xquote optTick dialog, and probably Danny's problem with
Xinvest, going.
If someone could post the results of form/test42.motif I can verify this.
Form.c 1.118
date: 1998/10/16 17:14:53; author: rwscott; state: Exp; lines: +20 -4
XmFormFindPreferred() - Instead of p = *cg only take those fields that are
valid according to the request_mode.
XmFormPath() - If the attachment is OPPOSITE_WIDGET add in the offset of the
widget that we are attached to. (messagebox/test11)
Form.c 1.117
date: 1998/10/16 01:38:50; author: rwscott; state: Exp; lines: +133 -309
I was adding in the margin width/height way to often. It is now only added
on when the attachment is ATTACH_FORM. This was messing up things with
attach_position and margins, like test35. Things still aren't perfect but,
they are alot closer. Motif must be using a "new math", because I can't
figure out where some of their numbers are coming from! We're mostly within
a couple of pixels of their values now.
A fair bit of clean-up also.
Frame.c Current rev 1.44 Last Release 1.43
Frame.c 1.44
date: 1998/10/13 16:09:58; author: rwscott; state: Exp; lines: +5 -5
A couple of 1 pixel tweaks to get all of the tests passing
GeoUtils.c Current rev 1.35 Last Release 1.28
GeoUtils.c 1.35
date: 1998/10/16 23:41:37; author: rwscott; state: Exp; lines: +23 -18
DebugUtil.c - debug stuff (no kidding)
BulletinBoard.c:query_geometry() - debug stuff
MainW.c:_XmMainWindowPreferredSize() - query the preferred size of the message
window (mainw/test2, 3, 4, 5, 6)
GeoUtils.c:_XmHandleQueryGeometry() - If policy is RESIZE_ANY we can't just
return the intended, we may have started this query off with a NULL
request to find the preferred size in the absence of any constraints.
This makes this look more like the query_geometry of Form.
(mainw/test2, 3, 4, 5, 6, messagebox/test3)
GeoUtils.c 1.34
date: 1998/10/04 01:54:50; author: rwscott; state: Exp; lines: +4 -4
In our continuing effort to go back in time.....
Rolled back MainW to 1.38. Not only was there the resize call in set_values
but many other things. Now at least most of the tests in mainw look half way
normal.
Rolled back ScrolledW to 1.87. Most of the changes since 1.88 have been
fixes for things that worked in 1.87!
This gets nedit looking correct again, Amaya displays a page again, etc ....
Now working
cascadebutton/test4
frame/test5
mainw/test1
mainw/test7
scrolledwindow/test16
scrolledwindow/test17
GeoUtils.c 1.33
date: 1998/09/30 03:41:26; author: rwscott; state: Exp; lines: +17 -2
MessageBox matrix create - Do not allocate a box for the symbol if it is of
DIALOG_MESSAGE or DIALOG_TEMPLATE type. This puts the symbol at +0+0 and the
message at the margins.
_XmGeoCalcFill() - When filling a row, the 2 boxes on the outside stay touching
the margin. All of the fill goes between the outside 2 boxes. In
other words, if we are filling a row that has 3 boxes we only have
2 areas of fill, not 4.
GeoUtils.c 1.32
date: 1998/09/30 01:36:22; author: rwscott; state: Exp; lines: +4 -2
Get the buttons in MessageBoxes the correct size. This is a third function
of the even_width/even_height member of the XmGeoRowLayoutRec. It holds an
increment to the width/height of the box. Since all of the buttons are 4
pixels too wide and narrow, if we load this value with 4 in the MatrixCreate
function we then add this to the height/width of each box during the
equal width/height processing.
Now to figure out how symbols are at +11+11 if they are there and +0+0 if
they are not there.
This makes some of the message box tests pass, and gets all of them closer
to a pass.
GeoUtils.c 1.31
date: 1998/09/29 00:21:07; author: rwscott; state: Exp; lines: +7 -2
GeoUtils.c:_XmGMReplyToQueryGeometry - add in a reply for RowColumn. Again
this will be removed to RowColumn if I can ever pick out a good
pattern to these things.
RowColumn.c:Initialize - set the width/height to 16 if not specified. test38
shows that this must be the case if we get a GeometryNo return.
RowColumn.c:query_geometry - don't let the preferred be less than 16 ??
And another wacky rule based on the request current and preferred (look
at the source, I can't explain this one in words)
This gets rowcolumn/test38 going and 36 and 39 are so close I can taste it. I
think there is something going on with borderWidth that I'm missing right now.
Xephem shows something rotten with borderWidth in RowColumns also, if I
remember correctly.
It's _dull_ Jon, but it has proven to be immensly useful.
GeoUtils.c 1.30
date: 1998/09/22 13:31:57; author: kbwj; state: Exp; lines: +8 -3
Added a DEBUGOUT statement to SegmentFill
GeoUtils.c 1.29
date: 1998/09/21 22:38:15; author: rwscott; state: Exp; lines: +7 -2
GeoUtils.c:_XmGMReplyToQueryGeometry() - add in the handler for ScrolledWindow
temporarily.
ScrolledW.c:query_geometry() - When trying to find the preferred size and
the width flag is not set, set the flag and use the GivenWidth. Ditto
for the height.
This is for the case of xmgrace. It was specifically setting the height of
the SW, so that is what it should end up being.
ImageCache.c Current rev 1.13 Last Release 1.11
ImageCache.c 1.13
date: 1998/10/13 05:48:20; author: gritton; state: Exp; lines: +8 -8
This time I fixed the "slant" pixmaps the right direction. Oops.
ImageCache.c 1.12
date: 1998/10/12 15:52:00; author: gritton; state: Exp; lines: +14 -14
Fixed builtin pixmaps to look like "that other implementation".
LTV6Editres.c Current rev 1.5 Last Release 1.4
LTV6Editres.c 1.5
date: 1998/10/21 19:06:49; author: danny; state: Exp; lines: +0 -3
Cosmetic patches by Alexander Mai.
Sorry for not applying them earlier, Alexander.
Mirrors.html : change site of German mirror
Label.c Current rev 1.80 Last Release 1.72
Label.c 1.80
date: 1998/09/28 01:35:16; author: rwscott; state: Exp; lines: +9 -3
query_geometry update shown by label tests 8 and 9
Label.c 1.79
date: 1998/09/25 03:22:23; author: rwscott; state: Exp; lines: +322 -193
The recent geometry changes in these were killing many other tests. A change
in Label does not just change Label! It changes PushB ToggleB and possibly
some things you don't even realize. For example the tests in just about every
directory use a label or labelgadget or one of its sub-classes. If you make
a change to something like this you had better have some damn good proof that
it is correct _and_ that it does not screw up three others that have been
working. Labels have been around for a _long_ time and have been picked apart
at the pixel level. Because of that they are used as the base for just about
every test in form. All of the form tests are also a good test of label since
they print out the geo of every widget. You make a change to something this
basic and you better be willing to back it up. I could not get through half
of my 30 or so test apps without something being messed up on the initial
screen, and most of these have been working for a _very_ long time.
Label.c 1.78
date: 1998/09/25 00:05:30; author: rwscott; state: Exp; lines: +14 -31
Label.c:query_geometry() - Who's been playing with this.
label/test5
cvs log Label.c
Are there any more??
nuff said.
Label.c 1.77
date: 1998/09/24 18:37:48; author: kbwj; state: Exp; lines: +37 -6
Fixed some problems width acceleratorText.
Label.c 1.76
date: 1998/09/23 04:21:51; author: kbwj; state: Exp; lines: +171 -314
Changed the cliprectangle, becourse it was too tight.
Label.c 1.75
date: 1998/09/22 18:15:06; author: jon; state: Exp; lines: +314 -175
Backing out a set of recent changes to restore some sanity till we kin
figger out what went wrong.
(The Texans are getting to me!)
Label.c 1.74
date: 1998/09/22 13:28:39; author: kbwj; state: Exp; lines: +177 -316
Added set_values_almost procedure, and changed set_values, resize and
expose to take advantage of it. Changed the size and placement of
cliprectangle.
Label.c 1.73
date: 1998/09/22 02:25:52; author: rwscott; state: Exp; lines: +6 -3
DragDrop.c:XmDragStart() - If the initiatorProtocolStyle is XmDRAG_NONE
return a NULL DragContext widget.
Label.c, LabelG.c, List.c - Deal with a NULL DragContext widget returned from
XmDragStart.
This allows you to turn off Dragging with the
*dragInitiatorProtocolStyle: XmDRAG_NONE
resource.
LabelG.c Current rev 1.67 Last Release 1.59
LabelG.c 1.67
date: 1998/09/28 01:35:18; author: rwscott; state: Exp; lines: +9 -3
query_geometry update shown by label tests 8 and 9
LabelG.c 1.66
date: 1998/09/25 03:22:24; author: rwscott; state: Exp; lines: +308 -211
The recent geometry changes in these were killing many other tests. A change
in Label does not just change Label! It changes PushB ToggleB and possibly
some things you don't even realize. For example the tests in just about every
directory use a label or labelgadget or one of its sub-classes. If you make
a change to something like this you had better have some damn good proof that
it is correct _and_ that it does not screw up three others that have been
working. Labels have been around for a _long_ time and have been picked apart
at the pixel level. Because of that they are used as the base for just about
every test in form. All of the form tests are also a good test of label since
they print out the geo of every widget. You make a change to something this
basic and you better be willing to back it up. I could not get through half
of my 30 or so test apps without something being messed up on the initial
screen, and most of these have been working for a _very_ long time.
LabelG.c 1.65
date: 1998/09/25 00:08:56; author: rwscott; state: Exp; lines: +14 -31
ditto
LabelG.c 1.64
date: 1998/09/24 18:37:49; author: kbwj; state: Exp; lines: +39 -7
Fixed some problems width acceleratorText.
LabelG.c 1.63
date: 1998/09/23 04:22:24; author: kbwj; state: Exp; lines: +184 -296
Changed call from _XmCalcLabelDimensions to _XmCalcLabelGDimensions
Changed the cliprectangle, becourse it was too tight.
LabelG.c 1.62
date: 1998/09/22 18:15:07; author: jon; state: Exp; lines: +294 -186
Backing out a set of recent changes to restore some sanity till we kin
figger out what went wrong.
(The Texans are getting to me!)
LabelG.c 1.61
date: 1998/09/22 13:28:40; author: kbwj; state: Exp; lines: +188 -296
Added set_values_almost procedure, and changed set_values, resize and
expose to take advantage of it. Changed the size and placement of
cliprectangle.
LabelG.c 1.60
date: 1998/09/22 02:25:53; author: rwscott; state: Exp; lines: +7 -3
DragDrop.c:XmDragStart() - If the initiatorProtocolStyle is XmDRAG_NONE
return a NULL DragContext widget.
Label.c, LabelG.c, List.c - Deal with a NULL DragContext widget returned from
XmDragStart.
This allows you to turn off Dragging with the
*dragInitiatorProtocolStyle: XmDRAG_NONE
resource.
List.c Current rev 1.66 Last Release 1.61
List.c 1.66
date: 1998/10/17 20:03:38; author: rwscott; state: Exp; lines: +29 -5
DrawingArea.c:change_managed() - I we are not realized and the width/height
is not zero use that instead of the calculated size. This allows
setting the size in a resource. (rowcolumn/test20)
_XmListSetSBManageHoriz() - if the SizePolicy is CONSTANT we were managing the
scrollbar without paying attention to the SBDisplayPolicy.
This now gets the correct widgets managed for many of the mainw, selectionbox
and command tests.
List.c 1.65
date: 1998/09/30 23:37:47; author: rwscott; state: Exp; lines: +36 -36
Swap the order that the scrollbars are created. Helps in geo comparisons
List.c 1.64
date: 1998/09/29 20:29:55; author: jon; state: Exp; lines: +7 -4
Plug leaks in this by moving ParseTranslationTable call to
class_initialize, and making list_trans a static variable. This brings
list/test17 from 500+ bytes leakage to 0. That's gotta help FSB, Rick.
Perhaps the funnel has a skinny end now.
-jon
List.c 1.63
date: 1998/09/22 02:25:54; author: rwscott; state: Exp; lines: +6 -3
DragDrop.c:XmDragStart() - If the initiatorProtocolStyle is XmDRAG_NONE
return a NULL DragContext widget.
Label.c, LabelG.c, List.c - Deal with a NULL DragContext widget returned from
XmDragStart.
This allows you to turn off Dragging with the
*dragInitiatorProtocolStyle: XmDRAG_NONE
resource.
List.c 1.62
date: 1998/09/20 00:38:51; author: rwscott; state: Exp; lines: +20 -2
List.c:ListDefaultAction() - If the parent of the list is a manager invoke
the parent process method.
List.c:ListKbdActivate() - simply call ListDefaultAction.
FileSB.c:initialize() - don't install the defaultAction callback on the file
list. ParentProcess is taking care of this now.
This gets rid of the double okCallback in the FileSB
MainW.c Current rev 1.43 Last Release 1.39
MainW.c 1.43
date: 1998/10/23 20:57:12; author: rwscott; state: Exp; lines: +20 -3
Move the stuff that changes the geometry request out of
_XmMainWindowGeomRequest and into change_managed. This way we really end up
requesting what we want to request.
MainW.c 1.42
date: 1998/10/23 17:27:40; author: rwscott; state: Exp; lines: +5 -2
BulletinBoard.c:change_managed() - Always do the layout, whether we are
realized or not.
MainW.c:realize() - Get rid of the geo stuff in here. It gets done in
change_managed, which is call just before realize.
ScrolledW.c:realize() - Just a comment for now. I don't think this geometry
stuff should be happening here.
This is all to get mainw/test13 working. It's not correct yet, the sizes are
still off but, at least it is put together correctly now.
MainW.c 1.41
date: 1998/10/16 23:41:39; author: rwscott; state: Exp; lines: +47 -14
DebugUtil.c - debug stuff (no kidding)
BulletinBoard.c:query_geometry() - debug stuff
MainW.c:_XmMainWindowPreferredSize() - query the preferred size of the message
window (mainw/test2, 3, 4, 5, 6)
GeoUtils.c:_XmHandleQueryGeometry() - If policy is RESIZE_ANY we can't just
return the intended, we may have started this query off with a NULL
request to find the preferred size in the absence of any constraints.
This makes this look more like the query_geometry of Form.
(mainw/test2, 3, 4, 5, 6, messagebox/test3)
MainW.c 1.40
date: 1998/10/04 01:54:52; author: rwscott; state: Exp; lines: +1542 -450
In our continuing effort to go back in time.....
Rolled back MainW to 1.38. Not only was there the resize call in set_values
but many other things. Now at least most of the tests in mainw look half way
normal.
Rolled back ScrolledW to 1.87. Most of the changes since 1.88 have been
fixes for things that worked in 1.87!
This gets nedit looking correct again, Amaya displays a page again, etc ....
Now working
cascadebutton/test4
frame/test5
mainw/test1
mainw/test7
scrolledwindow/test16
scrolledwindow/test17
Makefile.am Current rev 1.14 Last Release 1.13
Makefile.am 1.14
date: 1998/09/24 19:42:20; author: jon; state: Exp; lines: +2 -2
SGI's compiler has trouble with a gcc extension in xdnd.c:
#define dnd_debug(a,b...)
Apparently SGI's cpp can't handle the variable number of parameters.
Since we're not actually using xdnd at this point, I've taken xdnd out of
the build.
Manager.c Current rev 1.57 Last Release 1.54
Manager.c 1.57
date: 1998/10/18 22:01:00; author: rwscott; state: Exp; lines: +7 -2
DialogS.c - realize the shell in the initialize method instead of the
insert_child method.
Form.c - back out yesterdays change, since it was screwing up xephem, and the
real problem was in DialogS.
Manager.c - instrumentation
PanedW.c - instrumentation
This is wrt form/test43, 42 and Xquote.
The problem was that the DialogShell was being realized in its insert_child
method. This triggered a change_managed call to the paned window, which
would configure the form _before_ the changed_managed of form. Thus the
form children where the wrong size.
Manager.c 1.56
date: 1998/10/07 22:31:20; author: rwscott; state: Exp; lines: +5 -2
Manager.c:realize() - drawingarea/test7 shows that if the width/height is 0
then we bump it up to 1, to avoid X errors, but we do not set it
back to 0 after the XtCreateWindow.
DrawingArea.c:initialize() - drawingarea/test7 shows that we do not set the
size here if one is not specified.
DrawingArea.c:PreferedSize() - If we have no children, the preferred size
is our current size, if not 0 or the DA_MarginWidth if it is. (test7)
Manager.c 1.55
date: 1998/09/25 22:12:18; author: pgw; state: Exp; lines: +13 -2
Added code to initialize new 2.0 resource. Also, put preprocessor ifs around it to
support backward compatibility and to only include if building a 2.0 release.
MenuShell.c Current rev 1.70 Last Release 1.68
MenuShell.c 1.70
date: 1998/10/16 14:43:24; author: rwscott; state: Exp; lines: +24 -2
initialize() - copy the MS_DefaultFontList if it is specified.
xdir was segfaulting, when closing a connection, trying to free the
DefaultFontList.
MenuShell.c 1.69
date: 1998/09/20 23:04:31; author: jon; state: Exp; lines: +9 -4
I have ifdef'd out a test here which aborts MenuShellPopdownDone if the
triggering event is within the milticlick interval. Danny comments
"Hopefully this is an event we can discard". Apparently it's not in many
cases. I'm not sure why this test was put in anyway; perhaps Danny can
tell us when he returns.
This doesn't appear to break anything, but just in case, I've ifdef 0'd
the code instead of removing it in case we find a reason for it later. If
Danny decides to bless the change, he can commit a new version which
removes the dead section.
This fixes unreleased grabs with Mozilla and menushell/test7. If you wish
to demonstrate the bug, re-enable the ifdef 0 around line 660, and run
test7. Push button3 and release quiclky while moving the mouse left in
sort of a swoop motion. Don't try this unless you have another terminal
handy to kill the program or you ran the test from a shell script which
will automatically kill it.
MenuUtil.c Current rev 1.32 Last Release 1.30
MenuUtil.c 1.32
date: 1998/10/17 06:25:04; author: gritton; state: Exp; lines: +4 -4
Use _XmSleep for seconds, not _XmMicroSleep.
MenuUtil.c 1.31
date: 1998/10/08 05:23:56; author: gritton; state: Exp; lines: +12 -4
XmSetMenuCursor should set the cursor for all screens of the display.
MessageB.c Current rev 1.28 Last Release 1.24
MessageB.c 1.28
date: 1998/09/30 03:41:28; author: rwscott; state: Exp; lines: +6 -3
MessageBox matrix create - Do not allocate a box for the symbol if it is of
DIALOG_MESSAGE or DIALOG_TEMPLATE type. This puts the symbol at +0+0 and the
message at the margins.
_XmGeoCalcFill() - When filling a row, the 2 boxes on the outside stay touching
the margin. All of the fill goes between the outside 2 boxes. In
other words, if we are filling a row that has 3 boxes we only have
2 areas of fill, not 4.
MessageB.c 1.27
date: 1998/09/30 01:36:25; author: rwscott; state: Exp; lines: +12 -12
Get the buttons in MessageBoxes the correct size. This is a third function
of the even_width/even_height member of the XmGeoRowLayoutRec. It holds an
increment to the width/height of the box. Since all of the buttons are 4
pixels too wide and narrow, if we load this value with 4 in the MatrixCreate
function we then add this to the height/width of each box during the
equal width/height processing.
Now to figure out how symbols are at +11+11 if they are there and +0+0 if
they are not there.
This makes some of the message box tests pass, and gets all of them closer
to a pass.
MessageB.c 1.26
date: 1998/09/27 12:29:27; author: rwscott; state: Exp; lines: +46 -47
Change the order the children are created. This makes it much easier to
compare geometries
MessageB.c 1.25
date: 1998/09/23 04:24:23; author: kbwj; state: Exp; lines: +14 -2
Removed some air around the question bitmap, to fix geometry problems
in e.g messagebox/test2
PanedW.c Current rev 1.42 Last Release 1.39
PanedW.c 1.42
date: 1998/10/18 22:01:01; author: rwscott; state: Exp; lines: +30 -3
DialogS.c - realize the shell in the initialize method instead of the
insert_child method.
Form.c - back out yesterdays change, since it was screwing up xephem, and the
real problem was in DialogS.
Manager.c - instrumentation
PanedW.c - instrumentation
This is wrt form/test43, 42 and Xquote.
The problem was that the DialogShell was being realized in its insert_child
method. This triggered a change_managed call to the paned window, which
would configure the form _before_ the changed_managed of form. Thus the
form children where the wrong size.
PanedW.c 1.41
date: 1998/10/17 13:28:27; author: rwscott; state: Exp; lines: +13 -4
PanedW.c:realize() - Take out the geo negotiations, this will get done in
change_managed.
Form.c:XmFormFindPreferred() - Do not set the FCP_Pref[W|H] in here! Since
RC is initializing with a size of 16x16 now, Form was taking this
as a preferred size and laying out to this! (form/test42)
This gets the Xquote optTick dialog, and probably Danny's problem with
Xinvest, going.
If someone could post the results of form/test42.motif I can verify this.
PanedW.c 1.40
date: 1998/10/03 01:14:38; author: rwscott; state: Exp; lines: +6 -2
PanedW.c:destroy() - XtFree the PW_ManagedChildren list
Scale.c:destroy() - Free the font list.
ScrolledW.c:expose() - Only draw the shadows if we have a work window.
Text.c:Destroy() - We seem to have to XtDestroy the text.inner_widget also.
Why is Xt not taking care of this for us??
Primitive.c Current rev 1.34 Last Release 1.31
Primitive.c 1.34
date: 1998/10/10 07:38:59; author: gritton; state: Exp; lines: +3 -3
Fixed pixmap background color in _XmPrimitiveTopShadowPixmapDefault.
Primitive.c 1.33
date: 1998/10/06 18:09:16; author: pgw; state: Exp; lines: +29 -2
Added code to initialize new 2.0 resources only if we are building a 2.0
compatible version.
Primitive.c 1.32
date: 1998/09/22 13:29:13; author: kbwj; state: Exp; lines: +8 -5
fixed _XmDifferentBackground
PushB.c Current rev 1.48 Last Release 1.44
PushB.c 1.48
date: 1998/09/25 03:22:26; author: rwscott; state: Exp; lines: +109 -256
The recent geometry changes in these were killing many other tests. A change
in Label does not just change Label! It changes PushB ToggleB and possibly
some things you don't even realize. For example the tests in just about every
directory use a label or labelgadget or one of its sub-classes. If you make
a change to something like this you had better have some damn good proof that
it is correct _and_ that it does not screw up three others that have been
working. Labels have been around for a _long_ time and have been picked apart
at the pixel level. Because of that they are used as the base for just about
every test in form. All of the form tests are also a good test of label since
they print out the geo of every widget. You make a change to something this
basic and you better be willing to back it up. I could not get through half
of my 30 or so test apps without something being messed up on the initial
screen, and most of these have been working for a _very_ long time.
PushB.c 1.47
date: 1998/09/23 04:23:07; author: kbwj; state: Exp; lines: +256 -109
Changed default size when showAsDefault > 0 or
defalultbuttonShadowThickness > 0
PushB.c 1.46
date: 1998/09/22 18:15:07; author: jon; state: Exp; lines: +107 -254
Backing out a set of recent changes to restore some sanity till we kin
figger out what went wrong.
(The Texans are getting to me!)
PushB.c 1.45
date: 1998/09/22 13:29:45; author: kbwj; state: Exp; lines: +256 -109
Changed the initial size when defaultButtonShadowThickness > 0 or
showAsDefault > 0. Changed the color outside the shadow, to match the
parent when drawn as default
PushBG.c Current rev 1.52 Last Release 1.48
PushBG.c 1.52
date: 1998/09/25 03:22:26; author: rwscott; state: Exp; lines: +65 -209
The recent geometry changes in these were killing many other tests. A change
in Label does not just change Label! It changes PushB ToggleB and possibly
some things you don't even realize. For example the tests in just about every
directory use a label or labelgadget or one of its sub-classes. If you make
a change to something like this you had better have some damn good proof that
it is correct _and_ that it does not screw up three others that have been
working. Labels have been around for a _long_ time and have been picked apart
at the pixel level. Because of that they are used as the base for just about
every test in form. All of the form tests are also a good test of label since
they print out the geo of every widget. You make a change to something this
basic and you better be willing to back it up. I could not get through half
of my 30 or so test apps without something being messed up on the initial
screen, and most of these have been working for a _very_ long time.
PushBG.c 1.51
date: 1998/09/23 04:23:06; author: kbwj; state: Exp; lines: +209 -65
Changed default size when showAsDefault > 0 or
defalultbuttonShadowThickness > 0
PushBG.c 1.50
date: 1998/09/22 18:15:08; author: jon; state: Exp; lines: +63 -207
Backing out a set of recent changes to restore some sanity till we kin
figger out what went wrong.
(The Texans are getting to me!)
PushBG.c 1.49
date: 1998/09/22 13:30:14; author: kbwj; state: Exp; lines: +209 -65
Changed the initial size when defaultButtonShadowThickness > 0 or
showAsDefault > 0.
RCUtils.c Current rev 1.90 Last Release 1.85
RCUtils.c 1.90
date: 1998/10/26 21:38:37; author: rwscott; state: Exp; lines: +24 -4
PreferredSizeVT() - If we are being resized then we can't be any wider than
the RC. In this case split the available width up evenly between all
of the columns. This, and the next if, should maybe be done in the
Layout portion, since this is a constraint affecting the "preferred"
size. Then again, it is the preferred size given the constraints????
(rowcolumn/test42, xephem, xmgrace)
RCUtils.c 1.89
date: 1998/10/22 01:35:39; author: rwscott; state: Exp; lines: +4 -4
RowColumn.c - debug stuff
RCUtils.c:DoLayoutHC() - PreferredSizeHC and DoLayoutHC had a different idea
of the relationship between the width of the RC and the individual
widths of the children. DoLayout was simply using the width of the
RC divided by the number of kids in the row, totally ignoring spacing,
margins and so forth.
This, I think, gets Xquote performing as planned.
RCUtils.c 1.88
date: 1998/10/04 02:58:23; author: rwscott; state: Exp; lines: +8 -5
Small 5 pixel tweak to the positioning of the menu row when it gets wrapped.
Move the help widget a little more away from the right margin.
This get mainw/test14 perfect, both before and after the wrap. test12 is
close, but there are a lot of other widgets in that one.
RCUtils.c 1.87
date: 1998/09/28 01:44:42; author: rwscott; state: Exp; lines: +9 -4
_XmRCAdjustSize - get rid of some of the IsRealized !IsRealized baggage.
rowcolumn tests 1 and 25
RCUtils.c 1.86
date: 1998/09/22 13:30:47; author: kbwj; state: Exp; lines: +73 -2
Changed _XmRCGetKidGeo to deal with the size of CascadeButton when in
option menu
RepType.c Current rev 1.8 Last Release 1.6
RepType.c 1.8
date: 1998/10/16 04:49:28; author: gritton; state: Exp; lines: +216 -149
Since we know that rep_types[i].rep_type_id == i, there's no need to call
XmRepTypeGetRecord for internal use. It causes unnecessary malloc/free,
XmRepTypeAddReverse needs the uncopied record so it can modify it,
and some functions weren't freeing the copy.
XmRepTypeGetNameList shouldn't allocate its strings separately - it's
actually a single malloc'd chunk consisting of the array followed by the
strings. Thus the list is all that needs to be freed (and indeed all that
can be freed without a core dump). As I had to totally redo
__XmRepTypeCopyList, and it was only used in XmRepTypeGetNameList, I just
got rid of the static function and absorbed it.
The same is true with the copied records from XmRepTypeGetRecord and
XmRepTypeGetRegistered. So I added a static function __XmRepTypeCopyRecord
to handle this; XmRepTypeRegister also uses it (might as well).
XmRepTypesGetRegistered won't call XmRegisterConverters if I registered my
own stuff first (see comment).
Only warn in XmRepTypeValidValue is enable_default_warning is non-NULL.
A touch more checking in the converter functions.
RepType.c 1.7
date: 1998/09/30 06:27:56; author: gritton; state: Exp; lines: +50 -4
Added XmCvtStringToUnitType. It's here instead of ResConvert.c
because it wants to use __XmRepTypeStringToValue.
XmRepTypeGetRegistered allocates a list based on number_of_types,
but may then register types without growing the list accordingly.
ResConvert.c Current rev 1.39 Last Release 1.35
ResConvert.c 1.39
date: 1998/10/20 20:44:55; author: danny; state: Exp; lines: +7 -3
ResConvert : add support for True and False values for IndicatorOn.
The 2.0 change broke some stuff without this.
Toggles : a little more work. They toggle correctly at first sight,
now for the visual appearance ...
ResConvert.c 1.38
date: 1998/10/07 15:40:16; author: pgw; state: Exp; lines: +25 -2
ind_on resource in the ToggleButton gadget and widget was changed
in 2.0 tree from Boolean to unsigned char. This change was needed to
fix the following resource conversion problem:
Warning: Cannot convert string "INDICATOR_CHECK_BOX" to type Boolean
This change also breaks binary compatibility in the 2.0 tree.
ResConvert.c 1.37
date: 1998/10/04 07:23:15; author: gritton; state: Exp; lines: +394 -133
Added a new general converter from strings to comma-separated lists.
This handles the types: AtomList ButtonType[Table] CharSetTable KeySymTable
StringTable XmStringTable. Note XmStringTable now no longer has its own
converter, and its part of the generic function uses XmStringCreateLtoR
instead of XmStringCreateSimple. The table conversions have destructors,
and I've added one for XmFontList as well (which incedentally barely
escaped assimilation into the generic list converter).
Added a couple more simple string-to-int converters: TopItemPosition and
Window. That completes the list of "missing" converters/destructors.
ResConvert.c 1.36
date: 1998/09/30 06:29:21; author: gritton; state: Exp; lines: +304 -379
Almost all _XmCvt* functions are static (only the stuff in XmP.h isn't).
XmCvtStringToUnitType now exists, but in RepType.c
_XmCvtStringToFontlist changed to _XmCvtStringToXmFontList,
and it now specifies XtCacheByDisplay.
The obsolete _XmCvtStringToXmNavigation #ifdef'd out.
Added a couple of simple converters: Char and TextPosition.
Horizontal/Vertical converters have been unified into single
converters that understand ResInd stuff (calls _XmConvertUnits).
ResInd.c Current rev 1.9 Last Release 1.6
ResInd.c 1.9
date: 1998/10/08 05:24:51; author: gritton; state: Exp; lines: +20 -2
Filled in XmSetFontUnit and XmSetFontUnits.
ResInd.c 1.8
date: 1998/10/05 06:02:46; author: gritton; state: Exp; lines: +77 -132
New implementation of _XmConvertUnits. It uses all-integer arithmetic
instead of floating point. This works by doing all multiplies before all
divides. There is no rounding up by .5, as testing shows this doesn't
occur. All conversion are now performed, including the font-unit ones
that didn't exist before.
ResInd.c 1.7
date: 1998/09/30 17:07:02; author: gritton; state: Exp; lines: +6 -4
Only call _XmGetFontUnit if converting to or from Xm100TH_FONT_UNITS.
This breaks an infinite recursion caused when the screen object (and thus
the display object) is needed for the font units, but the font units are
needed for the display object's resource conversions. This works since
_XmGetUnitType always returns XmPIXELS for the display object shell.
What a silly cause for a crash - the results of _XmGetFontUnit aren't
even used yet.
RowColumn.c Current rev 1.173 Last Release 1.166
RowColumn.c 1.173
date: 1998/10/23 00:22:02; author: rwscott; state: Exp; lines: +27 -6
Whenever we change the labelString of the option menu button we also need to
change the labelType, labelPixmap and, labelInsensitivePixmap.
(rowcolumn/test41)
RowColumn.c 1.172
date: 1998/10/22 01:35:40; author: rwscott; state: Exp; lines: +36 -7
RowColumn.c - debug stuff
RCUtils.c:DoLayoutHC() - PreferredSizeHC and DoLayoutHC had a different idea
of the relationship between the width of the RC and the individual
widths of the children. DoLayout was simply using the width of the
RC divided by the number of kids in the row, totally ignoring spacing,
margins and so forth.
This, I think, gets Xquote performing as planned.
RowColumn.c 1.171
date: 1998/10/16 21:01:43; author: pgw; state: Exp; lines: +20 -2
Commit changes to handle the XmNpopupEnabled 2.0 feature.
RowColumn.c 1.170
date: 1998/09/29 15:04:07; author: rwscott; state: Exp; lines: +4 -4
query_geometry - Small tweak using this magic 16 that keeps coming up.
This gets test36 and 39 going.
RowColumn.c 1.169
date: 1998/09/29 00:21:08; author: rwscott; state: Exp; lines: +22 -2
GeoUtils.c:_XmGMReplyToQueryGeometry - add in a reply for RowColumn. Again
this will be removed to RowColumn if I can ever pick out a good
pattern to these things.
RowColumn.c:Initialize - set the width/height to 16 if not specified. test38
shows that this must be the case if we get a GeometryNo return.
RowColumn.c:query_geometry - don't let the preferred be less than 16 ??
And another wacky rule based on the request current and preferred (look
at the source, I can't explain this one in words)
This gets rowcolumn/test38 going and 36 and 39 are so close I can taste it. I
think there is something going on with borderWidth that I'm missing right now.
Xephem shows something rotten with borderWidth in RowColumns also, if I
remember correctly.
It's _dull_ Jon, but it has proven to be immensly useful.
RowColumn.c 1.168
date: 1998/09/22 13:31:24; author: kbwj; state: Exp; lines: +19 -16
Moved the mnemonic to the OptionLabel Widget
RowColumn.c 1.167
date: 1998/09/21 01:08:07; author: rwscott; state: Exp; lines: +33 -25
_XmMenuFocus() - XmMENU_FOCUS_RESTORE - Sometimes torn off pop-up menu
selections are causing a segfault because oldWidget == NULL. This
means that somewhere we are missing an XmMENU_FOCUS_SAVE. For now
just detect the problem and print a big ole warning so I won't forget
about it.
Scale.c Current rev 1.30 Last Release 1.28
Scale.c 1.30
date: 1998/10/24 00:38:14; author: rwscott; state: Exp; lines: +43 -6
The pixel police stopped by Scale and placed a couple of pixel shims between
some of the elements. This gets all the scale tests working.
Also get rid of the geo negotiation from initialize and realize. This will
get done in change_managed.
Scale.c 1.29
date: 1998/10/03 01:14:40; author: rwscott; state: Exp; lines: +3 -4
PanedW.c:destroy() - XtFree the PW_ManagedChildren list
Scale.c:destroy() - Free the font list.
ScrolledW.c:expose() - Only draw the shadows if we have a work window.
Text.c:Destroy() - We seem to have to XtDestroy the text.inner_widget also.
Why is Xt not taking care of this for us??
Screen.c Current rev 1.15 Last Release 1.11
Screen.c 1.15
date: 1998/10/14 21:38:26; author: rwscott; state: Exp; lines: +11 -2
Add some instrumentation to Screen to see if we keep allocating new ones. I
can't see an infinite loop here. Everything I've run creates only 1 screen.
We need more details, a test case, or at least the output from the app with
DEBUGSOURCES=Screen.c, after this update.
Screen.c 1.14
date: 1998/10/08 05:32:06; author: gritton; state: Exp; lines: +83 -83
Delete the screen-object context mapping in the destroy method, and
invalidate the color cache.
Broke the function _XmComputeFontUnits called from initialize into the
call-procs default_horizontal_font_unit and default_vertical_font_unit.
This allows for loading the resources with initial values.
These call-procs aren't quite like _XmComputeFontUnits: they use
XGetFontProperty insread of peeking directly into the font struct,
use integer arithmetic, and round off the division (to make up for when
I made _XmConvertUnits stop rounding :-).
In set_values, changing the font changes the font units, unless they were
explicitly changed at the same time. set_values also checks the
unpostBehavior resource with XmRepTypeValidValue.
Screen.c 1.13
date: 1998/10/02 14:57:46; author: rwscott; state: Exp; lines: +35 -13
Screen.c - add some instrumentation to Alloc/Free pixmap
CascadeB[G] - make them use the pixmap cache for both allocating and freeing
the arrow pixmaps. Make sure the pixmap instance variables get
initialized.
Screen.c 1.12
date: 1998/09/21 16:40:48; author: jon; state: Exp; lines: +7 -10
Checking in Jamie's "final" color patches. Hummm, where have I heard that
one before?
This is probably my last addition to Visual.c's color handling. With this
patch everything listed in XmP.h is implemented, except for
_XmGetDefaultThresholdsForScreen. This one stumps me. The header file
shows it as a CallProc for a dynamic resource default. Yet despite the
docs for the Screen object, those values aren't dynamic at all. They're
zero - as undynamic as you can get. In HP-UX's Motif 1.2 headers, the
function is listed as taking only a Screen* argument. But if I call it,
using either calling convention, nothing changes (as far as I can tell),
certainly not the Screen object's threshold resources, and no XrmValue is
returned. But I digress. Here's the summary of what this patch *does*
do:
_XmColorProcDefaultProc uses threshold resource values from the Screen
object if they're not the default (0).
Changing color proc or Screen thresholds invalidates the color cache.
The fallback to black or white on a failed color allocation is "smarter",
including generally making foreground and select colors different.
_XmBackgroundColorDefault uses the same color cache as everything else.
Added _XmGetDefaultColors, _XmGetDefaultBackgroundColorSpec (used by
_XmBackgroundColorDefault), and _XmSetDefaultBackgroundColorSpec.
ScrollBar.c Current rev 1.36 Last Release 1.35
ScrollBar.c 1.36
date: 1998/10/10 07:40:45; author: gritton; state: Exp; lines: +6 -6
Fix _XmDrawArrow for proper shadow thickness on 2.0.
ArrowButton: shadow should always be 2.
ScrollBar: shadow should be 0, 1, or 2.
ScrolledW.c Current rev 1.100 Last Release 1.89
ScrolledW.c 1.100
date: 1998/10/24 03:18:41; author: rwscott; state: Exp; lines: +6 -4
ScrolledW.c:change_managed() - Always do the geo negotiation.
ScrolledW.c:_XmConfigureScrollBars() - Do not un-manage the scrollbars if the
work window is not managed. Remember, we are just pushing them off
the side now.
ScrolledW.c:_XmScrolledWConfigureChildren() - Do not configure an un-managed
child. This screws up the initial sizing of said child.
Text.c - add a new-line to a debug statement
TextOut.c - add a new-line to a debug statement
This gets <textarea rows=y cols=x></textarea> going in Mozilla and,
<select size=5><option></select>
Which means that the bug pages finally look correct!!
ScrolledW.c 1.99
date: 1998/10/23 17:27:42; author: rwscott; state: Exp; lines: +8 -2
BulletinBoard.c:change_managed() - Always do the layout, whether we are
realized or not.
MainW.c:realize() - Get rid of the geo stuff in here. It gets done in
change_managed, which is call just before realize.
ScrolledW.c:realize() - Just a comment for now. I don't think this geometry
stuff should be happening here.
This is all to get mainw/test13 working. It's not correct yet, the sizes are
still off but, at least it is put together correctly now.
ScrolledW.c 1.98
date: 1998/10/23 05:15:51; author: rwscott; state: Exp; lines: +18 -3
ScrolledW.c:_XmScrolledWPreferredSize() - If the VisualPolicy is constant
and the width/height is 0, use the default of 100 instead of 0.
(scrolledwindow/test6, 10)
ScrolledW.c:_XmScrolledWGeomRequest() - This has bugged me for a while.
list/test15 shows that this crap should not be here. This also makes sense
since if we call _XmScrolledWGeomRequest with a request we should not be
altering it! test15 shows that we are getting in here with the correct
values, but this is changing them. If they need to be changed in some
situations _this is not the place to do it_.
SelectionBox.c:XmCreateSelectionDialog() - use XmCreateDialogShell to get the
correct widget tree.
ScrolledW.c 1.97
date: 1998/10/20 19:41:37; author: rwscott; state: Exp; lines: +5 -3
scrolledwindow/test11 was not resizing the drawing area correctly. This is
a scrolledwindow in AUTOMATIC mode. This change gets test11 working, but
needs a little more looking into. I want to pick it up at another location,
so I am committing it now.
ScrolledW.c 1.96
date: 1998/10/13 00:43:12; author: rwscott; state: Exp; lines: +28 -10
Add some ScrolledW instrumentation
Don't un-manage the scrollbar if it should not be displayed, just push it off
the edge (scrolledwindow/test10).
Minor adjustments to the size and position of the clipWindow.
This gets about 4 more of the scrolledwindow tests passing.
ScrolledW.c 1.95
date: 1998/10/05 21:36:41; author: rwscott; state: Exp; lines: +18 -18
initialize() - Make the clipWindow before the scroll bars. Make it able
to compare the geos.
It seems that our clip window is 2 pixels too heigh/wide which may be causing
the scrollbars to be too big.
ScrolledW.c 1.94
date: 1998/10/04 01:54:53; author: rwscott; state: Exp; lines: +1581 -1988
In our continuing effort to go back in time.....
Rolled back MainW to 1.38. Not only was there the resize call in set_values
but many other things. Now at least most of the tests in mainw look half way
normal.
Rolled back ScrolledW to 1.87. Most of the changes since 1.88 have been
fixes for things that worked in 1.87!
This gets nedit looking correct again, Amaya displays a page again, etc ....
Now working
cascadebutton/test4
frame/test5
mainw/test1
mainw/test7
scrolledwindow/test16
scrolledwindow/test17
ScrolledW.c 1.93
date: 1998/10/03 01:14:42; author: rwscott; state: Exp; lines: +23 -20
PanedW.c:destroy() - XtFree the PW_ManagedChildren list
Scale.c:destroy() - Free the font list.
ScrolledW.c:expose() - Only draw the shadows if we have a work window.
Text.c:Destroy() - We seem to have to XtDestroy the text.inner_widget also.
Why is Xt not taking care of this for us??
ScrolledW.c 1.92
date: 1998/09/24 19:05:25; author: jon; state: Exp; lines: +4 -2
SGI's compiler apparently doesn't like empty statements in switch
statements, e.g.:
switch (foo) {
case bar;
baz++;
break;
default:
}
So I inserted NULL; after default. And there was much rejoicing.
ScrolledW.c 1.91
date: 1998/09/21 22:48:32; author: rwscott; state: Exp; lines: +7 -3
_XmInitializeScrollBars() - clean out some unused baggage.
ScrolledW.c 1.90
date: 1998/09/21 22:38:18; author: rwscott; state: Exp; lines: +31 -2
GeoUtils.c:_XmGMReplyToQueryGeometry() - add in the handler for ScrolledWindow
temporarily.
ScrolledW.c:query_geometry() - When trying to find the preferred size and
the width flag is not set, set the flag and use the GivenWidth. Ditto
for the height.
This is for the case of xmgrace. It was specifically setting the height of
the SW, so that is what it should end up being.
SelectionBox.c Current rev 1.37 Last Release 1.34
SelectionBox.c 1.37
date: 1998/10/23 05:15:52; author: rwscott; state: Exp; lines: +6 -2
ScrolledW.c:_XmScrolledWPreferredSize() - If the VisualPolicy is constant
and the width/height is 0, use the default of 100 instead of 0.
(scrolledwindow/test6, 10)
ScrolledW.c:_XmScrolledWGeomRequest() - This has bugged me for a while.
list/test15 shows that this crap should not be here. This also makes sense
since if we call _XmScrolledWGeomRequest with a request we should not be
altering it! test15 shows that we are getting in here with the correct
values, but this is changing them. If they need to be changed in some
situations _this is not the place to do it_.
SelectionBox.c:XmCreateSelectionDialog() - use XmCreateDialogShell to get the
correct widget tree.
SelectionBox.c 1.36
date: 1998/10/02 17:56:31; author: rwscott; state: Exp; lines: +12 -2
Command.c:initialize() - Since Com_PromptString and ComCommand are just
re-named SelectionBox resources, let SelectionBox deal with them.
Command.c:set_values() - ditto
Command.c:destroy() - ditto
SelectionBox.c:initialize() - if the textString resource is specified make sure
it gets stuffed into the TextField
I suspect that we have a few of these kick'n around. Basically if a class
uses one of its superclasses resources with a name change, the super-class
will take care of anything to do with that resource.
SelectionBox.c 1.35
date: 1998/09/20 00:51:48; author: rwscott; state: Exp; lines: +4 -2
FileSB.c:initialize() - No need to remove all the default action callbacks
now since SelectionBox is not adding one.
SelectionBox.c:_XmSelectionBoxCreateList() - Don't add a defaultAction callback
to the list. This is now handled with ParentProcess.
ShadowI.c Current rev 1.5 Last Release 1.4
ShadowI.c 1.5
date: 1998/10/10 07:42:31; author: gritton; state: Exp; lines: +8 -11
Fixed _XmDrawDiamond_1_2_ to use proper semantics for "fill" argument.
It's not whether to fill, but whether to fill all the way to the
diamond shadows (as opposed to stopping one pixel short).
Simple.c Current rev 1.15 Last Release 1.14
Simple.c 1.15
date: 1998/10/04 07:23:16; author: gritton; state: Exp; lines: +3 -3
Added a new general converter from strings to comma-separated lists.
This handles the types: AtomList ButtonType[Table] CharSetTable KeySymTable
StringTable XmStringTable. Note XmStringTable now no longer has its own
converter, and its part of the generic function uses XmStringCreateLtoR
instead of XmStringCreateSimple. The table conversions have destructors,
and I've added one for XmFontList as well (which incedentally barely
escaped assimilation into the generic list converter).
Added a couple more simple string-to-int converters: TopItemPosition and
Window. That completes the list of "missing" converters/destructors.
Text.c Current rev 1.72 Last Release 1.68
Text.c 1.72
date: 1998/10/24 03:18:42; author: rwscott; state: Exp; lines: +3 -3
ScrolledW.c:change_managed() - Always do the geo negotiation.
ScrolledW.c:_XmConfigureScrollBars() - Do not un-manage the scrollbars if the
work window is not managed. Remember, we are just pushing them off
the side now.
ScrolledW.c:_XmScrolledWConfigureChildren() - Do not configure an un-managed
child. This screws up the initial sizing of said child.
Text.c - add a new-line to a debug statement
TextOut.c - add a new-line to a debug statement
This gets <textarea rows=y cols=x></textarea> going in Mozilla and,
<select size=5><option></select>
Which means that the bug pages finally look correct!!
Text.c 1.71
date: 1998/10/03 14:49:06; author: rwscott; state: Exp; lines: +9 -4
Trying to find out why we have lost nedit, among one of the first apps that
we ever got working.
Add some instrumentation in aid of the above goal. It turns out that it is
because of the massive change to MainW.c rev 1.39. Specifically the resize
call in the set_values method. We can suggest a new size in this method by
changing XtWidth/XtHeight _but_ we cannot force such a change by calling
resize. This not only lost nedit for us but _many_ of the tests in mainw!
Now to go through the diff from 1.38 to 1.39 and figure out what can stay and
what has to go. This may take awhile!
Text.c 1.70
date: 1998/10/03 01:14:44; author: rwscott; state: Exp; lines: +6 -2
PanedW.c:destroy() - XtFree the PW_ManagedChildren list
Scale.c:destroy() - Free the font list.
ScrolledW.c:expose() - Only draw the shadows if we have a work window.
Text.c:Destroy() - We seem to have to XtDestroy the text.inner_widget also.
Why is Xt not taking care of this for us??
Text.c 1.69
date: 1998/09/30 06:24:35; author: gritton; state: Exp; lines: +2 -5
Removed unused reference to the unused now-static _XmCvtStringToXmString.
TextF.c Current rev 1.71 Last Release 1.68
TextF.c 1.71
date: 1998/10/01 12:40:14; author: rwscott; state: Exp; lines: +6 -2
initialize() - If a FontList is specified here we must make a copy of it since
we will be freeing it in destroy.
This was showing up in ml->compose->new->cancel->yes.
TextF.c 1.70
date: 1998/09/30 06:26:20; author: gritton; state: Exp; lines: +2 -11
Removed call to XtSetTypeConverter( ...XmString...), which is already done
in _XmRegisterConverters. Besides, _XmCvtStringToXmString is static now.
TextF.c 1.69
date: 1998/09/29 20:59:12; author: jon; state: Exp; lines: +14 -7
Move translation table parsing to class initialize a la BulletinBoard.
Plugs leak in textf/test8 and selectionbox/test9.
TextOut.c Current rev 1.48 Last Release 1.47
TextOut.c 1.48
date: 1998/10/24 03:18:44; author: rwscott; state: Exp; lines: +3 -3
ScrolledW.c:change_managed() - Always do the geo negotiation.
ScrolledW.c:_XmConfigureScrollBars() - Do not un-manage the scrollbars if the
work window is not managed. Remember, we are just pushing them off
the side now.
ScrolledW.c:_XmScrolledWConfigureChildren() - Do not configure an un-managed
child. This screws up the initial sizing of said child.
Text.c - add a new-line to a debug statement
TextOut.c - add a new-line to a debug statement
This gets <textarea rows=y cols=x></textarea> going in Mozilla and,
<select size=5><option></select>
Which means that the bug pages finally look correct!!
ToggleB.c Current rev 1.35 Last Release 1.32
ToggleB.c 1.35
date: 1998/10/10 07:37:36; author: gritton; state: Exp; lines: +11 -7
The "select" inside of the indicator is shrunk by 1 pixel on B/W screens.
ToggleB.c 1.34
date: 1998/09/28 01:49:54; author: rwscott; state: Exp; lines: +4 -5
initialize() - We were decreasing the IndicatorDim for a ONE_OF_MANY for some
reason. rowcolumn tests 8 9 19 and 35 show this. Hopefully this will
be verified when I get the Expected values for togglebuttons. It's a
2 pixel increase in width for ONE_OF_MANY toggle buttons.
With this we now also have rowcolumn tests 1 13 19 25 31 35 4 and 8 passing.
That concludes programming for another day bzzzzzzz
ToggleB.c 1.33
date: 1998/09/27 03:13:21; author: rwscott; state: Exp; lines: +8 -2
I can get the toggles in the rowcolumn tests the right size but, this loses
label/test3. So just a comment for now until some good toggle tests are
done to figure this out.
ToggleBG.c Current rev 1.35 Last Release 1.31
ToggleBG.c 1.35
date: 1998/10/10 07:37:36; author: gritton; state: Exp; lines: +11 -7
The "select" inside of the indicator is shrunk by 1 pixel on B/W screens.
ToggleBG.c 1.34
date: 1998/10/07 15:40:17; author: pgw; state: Exp; lines: +19 -2
ind_on resource in the ToggleButton gadget and widget was changed
in 2.0 tree from Boolean to unsigned char. This change was needed to
fix the following resource conversion problem:
Warning: Cannot convert string "INDICATOR_CHECK_BOX" to type Boolean
This change also breaks binary compatibility in the 2.0 tree.
ToggleBG.c 1.33
date: 1998/09/28 01:49:55; author: rwscott; state: Exp; lines: +5 -4
initialize() - We were decreasing the IndicatorDim for a ONE_OF_MANY for some
reason. rowcolumn tests 8 9 19 and 35 show this. Hopefully this will
be verified when I get the Expected values for togglebuttons. It's a
2 pixel increase in width for ONE_OF_MANY toggle buttons.
With this we now also have rowcolumn tests 1 13 19 25 31 35 4 and 8 passing.
That concludes programming for another day bzzzzzzz
ToggleBG.c 1.32
date: 1998/09/27 03:13:22; author: rwscott; state: Exp; lines: +8 -3
I can get the toggles in the rowcolumn tests the right size but, this loses
label/test3. So just a comment for now until some good toggle tests are
done to figure this out.
Traversal.c Current rev 1.28 Last Release 1.27
Traversal.c 1.28
date: 1998/10/01 02:23:09; author: rwscott; state: Exp; lines: +3 -3
Searching for memory leaks can reveal some interesting things. Trying to
track down some of the FSB leaks led me to SB which led to TF which led to
Primitive. The TF is created with a sticky navigation type, so an entry gets
added onto the tab list. When we destroy the widget we go to take the entry
off the tab list but, the src and dest where reversed in the bcopy which
claims the empty spot after the removal.
For example after a selection box is created the tab list is as follows:
test9 ItemsListVSB ItemsListHSB Text
delete ItemsListVSB and we were getting
test9 ItemsListVSB ItemsListVSB
instead of
test9 ItemsListHSB Text
which all means when we went to take off ItemsListHSB or Text we could not
find it, and the tab list would just keep growing.
This doesn't help the FSB leak problem much but it should keep the traversal
straight, and clears up the selection box leak, and probably others as well.
Visual.c Current rev 1.24 Last Release 1.19
Visual.c 1.24
date: 1998/10/10 07:46:03; author: gritton; state: Exp; lines: +129 -57
On a B/W screen, the default background color is always white,
regardless of the background color spec.
On a B/W screen, don't call the color proc, but use a static arrangement
instead. Added the call_color_proc function to streamline this.
Changed _XmForegroundColorDefault etc slightly - no visible effect.
_XmWhitePixel and _XmBlackPixel now have a "mini-cache" for XAllocColor.
Visual.c 1.23
date: 1998/09/30 06:23:39; author: gritton; state: Exp; lines: +11 -11
Almost all _XmCvt* functions are static (only the stuff in XmP.h isn't).
Visual.c 1.22
date: 1998/09/24 14:46:33; author: gritton; state: Exp; lines: +3 -2
Added highlightColor to the resources set by XmChangeColor
Visual.c 1.21
date: 1998/09/22 22:53:53; author: gritton; state: Exp; lines: +120 -285
Mwm now uses libXm's (Visual.c) color calculation functions, via the
_XmAddBackgroundToColorCache and XmGetColors functions.
_XmAddBackgroundToColorCache is a non-Motif (internal) function which
does (most of) the color-cache part of _XmBackgroundColorDefault (which
now calls it).
Now that mwm no longer uses the non-Motif __XmRGB2HSV and __XmHSV2RGB
fnuctions (actually, it never used the latter), I'm taking them out.
They used to be in libXm's Visual.c as well as two places in mwm
(misc.c and commented out in defaults.c).
Also, acting on a tip from segg@reseau.com, I'm including an empty
_XmGetDefaultThresholdsForScreen function to Visual.c - what it apparently
does, I already do elsewhere. While I was at it, I changed some
XDisplayOfScreen functions to DisplayOfScreen macros, just because.
Visual.c 1.20
date: 1998/09/21 16:40:48; author: jon; state: Exp; lines: +298 -69
Checking in Jamie's "final" color patches. Hummm, where have I heard that
one before?
This is probably my last addition to Visual.c's color handling. With this
patch everything listed in XmP.h is implemented, except for
_XmGetDefaultThresholdsForScreen. This one stumps me. The header file
shows it as a CallProc for a dynamic resource default. Yet despite the
docs for the Screen object, those values aren't dynamic at all. They're
zero - as undynamic as you can get. In HP-UX's Motif 1.2 headers, the
function is listed as taking only a Screen* argument. But if I call it,
using either calling convention, nothing changes (as far as I can tell),
certainly not the Screen object's threshold resources, and no XrmValue is
returned. But I digress. Here's the summary of what this patch *does*
do:
_XmColorProcDefaultProc uses threshold resource values from the Screen
object if they're not the default (0).
Changing color proc or Screen thresholds invalidates the color cache.
The fallback to black or white on a failed color allocation is "smarter",
including generally making foreground and select colors different.
_XmBackgroundColorDefault uses the same color cache as everything else.
Added _XmGetDefaultColors, _XmGetDefaultBackgroundColorSpec (used by
_XmBackgroundColorDefault), and _XmSetDefaultBackgroundColorSpec.
XmString.c Current rev 1.32 Last Release 1.31
XmString.c 1.32
date: 1998/10/01 21:10:10; author: danny; state: Exp; lines: +12 -2
Additions by Frederick A Niles <niles@axp745.gsfc.nasa.gov> (Rick).
Xmos.c Current rev 1.23 Last Release 1.21
Xmos.c 1.23
date: 1998/10/17 23:40:38; author: gritton; state: Exp; lines: +38 -65
Use select() in _XmMicroSleep instead of Xpoll.h's Select(), as Rick
Scott couldn't find Xpoll.h, and it's not documented anywhere.
Include the heder-exists defines from LTconfig.h, since they're there.
More work on _XmOSGetHomeDirName: try: HOME, getpwnam(LOGNAME),
getpwnam(USER), getpwuid(getuid()), "". The last precludes returning NULL.
Since _XmOSGetHomeDirName doesn't return NULL, convertTwiddle doesn't
return NULL either any more.
Since _XmOSGetHomeDirName, convertTwiddle, and XtNewString never return
NULL, I can get rid of those abort() calls in _XmOSQualifyFileSpec.
Good: their presence bothered me.
Xmos.c 1.22
date: 1998/10/17 06:21:16; author: gritton; state: Exp; lines: +45 -18
_XmMicroSleep is indeed microseconds, but there's also _XmSleep for seconds.
_XmOSGetHomeDirName now caches its result, and checks the passwd file
if the HOME environment variable doesn't exist.
misc.c Current rev 1.10 Last Release 1.9
misc.c 1.10
date: 1998/10/21 22:46:34; author: danny; state: Exp; lines: +72 -1
Add a couple of internal API's to deal with multitheading.
Document in CURRENT_NOTES.
Still need someone to explain whether what I did to the Toggle sources
is good enough.
Executive Summary
Of 109 files, 54 changed, and 55 had no change
File Current Rev. Last Release
.indent.pro 1.1 1.1 No Change
ArrowB.c 1.24 1.22
ArrowBG.c 1.21 1.19
AtomMgr.c 1.4 1.4 No Change
BaseClass.c 1.10 1.10 No Change
BulletinBoard.c 1.38 1.32
Cache.c 1.8 1.8 No Change
CascadeB.c 1.96 1.87
CascadeBG.c 1.85 1.78
Command.c 1.14 1.11
CutPaste.c 1.7 1.7 No Change
DebugUtil.c 1.39 1.35
Desktop.c 1.5 1.5 No Change
Dest.c 1.2 1.2 No Change
DialogS.c 1.16 1.12
Display.c 1.13 1.13 No Change
DragBS.c 1.10 1.8
DragC.c 1.19 1.19 No Change
DragDrop.c 1.7 1.6
DragIcon.c 1.9 1.9 No Change
DragOverS.c 1.13 1.12
DragUnder.c 1.3 1.3 No Change
DrawingArea.c 1.30 1.26
DrawnB.c 1.29 1.26
DropSMgr.c 1.17 1.17 No Change
DropTrans.c 1.9 1.9 No Change
ExtObject.c 1.4 1.4 No Change
FileSB.c 1.59 1.55
FontList.c 1.14 1.14 No Change
Form.c 1.123 1.116
Frame.c 1.44 1.43
Gadget.c 1.12 1.12 No Change
GadgetUtil.c 1.8 1.8 No Change
GeoUtils.c 1.35 1.28
Hash.c 1.4 1.4 No Change
ImageCache.c 1.13 1.11
LTCvt.c 1.5 1.5 No Change
LTV5Editres.c 1.2 1.2 No Change
LTV6Editres.c 1.5 1.4
LTXpm.c 1.6 1.6 No Change
LTfnmatch.c 1.3 1.3 No Change
LTglob.c 1.6 1.6 No Change
Label.c 1.80 1.72
LabelG.c 1.67 1.59
List.c 1.66 1.61
MainW.c 1.43 1.39
Makefile.am 1.14 1.13
Manager.c 1.57 1.54
MapEvent.c 1.12 1.12 No Change
MenuShell.c 1.70 1.68
MenuUtil.c 1.32 1.30
MessageB.c 1.28 1.24
Messages.c 1.3 1.3 No Change
Mwm.c 1.3 1.3 No Change
Obsolete.c 1.3 1.3 No Change
PanedW.c 1.42 1.39
Primitive.c 1.34 1.31
Protocols.c 1.4 1.4 No Change
PushB.c 1.48 1.44
PushBG.c 1.52 1.48
RCUtils.c 1.90 1.85
ReadImage.c 1.2 1.2 No Change
Region.c 1.4 1.4 No Change
RepType.c 1.8 1.6
ResConvert.c 1.39 1.35
ResInd.c 1.9 1.6
RowColumn.c 1.173 1.166
Sash.c 1.12 1.12 No Change
Scale.c 1.30 1.28
Screen.c 1.15 1.11
ScrollBar.c 1.36 1.35
ScrolledW.c 1.100 1.89
Selection.c 1.4 1.4 No Change
SelectionBox.c 1.37 1.34
SeparatoG.c 1.11 1.11 No Change
Separator.c 1.11 1.11 No Change
Shadow.c 1.13 1.13 No Change
ShadowI.c 1.5 1.4
ShellExt.c 1.4 1.4 No Change
Simple.c 1.15 1.14
TearOff.c 1.27 1.27 No Change
TearOffB.c 1.16 1.16 No Change
Text.c 1.72 1.68
TextF.c 1.71 1.68
TextIn.c 1.34 1.34 No Change
TextOut.c 1.48 1.47
TextStrSo.c 1.19 1.19 No Change
ToggleB.c 1.35 1.32
ToggleBG.c 1.35 1.31
TrackLoc.c 1.4 1.4 No Change
Transltns.c 1.6 1.6 No Change
TravAct.c 1.8 1.8 No Change
Traversal.c 1.28 1.27
UniqueEvent.c 1.4 1.4 No Change
VaSimple.c 1.3 1.3 No Change
Vendor.c 1.38 1.38 No Change
VirtKeys.c 1.9 1.9 No Change
Visual.c 1.24 1.19
World.c 1.4 1.4 No Change
XmIm.c 1.8 1.8 No Change
XmOS2.def.dist 1.2 1.2 No Change
XmStrDefs.c 1.2 1.2 No Change
XmString.c 1.32 1.31
Xmos.c 1.23 1.21
alloca.c 1.2 1.2 No Change
inlines.in 1.3 1.3 No Change
misc.c 1.10 1.9
mkinline.c 1.4 1.4 No Change
xdnd.c 1.2 1.2 No Change
|