thdltools/Jskad
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
749 commits 1 branch 0 tags 10 MiB
Commit graph

332 commits

Author SHA1 Message Date
dchandler
7304c770c9 Just a better comment. 2005-02-05 16:27:34 +00:00
a1tsal
28d46bb207 Add corrective comment regarding the bogus Unicode OM characters. 2005-01-19 01:07:52 +00:00
a1tsal
affccad9e5 Use 00A0 rather than 0020 for _, per unicode spec. 2005-01-17 08:58:56 +00:00
a1tsal
91d0e7f4da Use "precomposed" sanskrit consonant combinations consistently throughout. 2005-01-17 08:49:04 +00:00
a1tsal
22cfee69db Had wrong Unicode for n+n+y. 
Had wrong Unicode for space -- but only in comment.

Cleaned up punctuation in another comment.
 2005-01-16 01:17:44 +00:00
dchandler
0b0af67ed9 Ximalaya is not nearly as nice as Tibetan Machine Uni, so use the latter. 2005-01-04 02:20:59 +00:00
dchandler
aa5d86a6e3 The *->Unicode conversions were outputting Unicode that was not 
well-formed.  They still do, but they do it less often.

Chris Fynn wrote this a while back:

   By normal Tibetan & Dzongkha spelling, writing, and input rules
   Tibetan script stacks should be entered and written: 1 headline
   consonant (0F40->0F6A), any subjoined consonant(s) (0F90-> 0F9C),
   achung (0F71), shabkyu (0F74), any above headline vowel(s) (0F72
   0F7A 0F7B 0F7C 0F7D and 0F80); any ngaro (0F7E, 0F82 and 0F83).

Now efforts are made to ensure that the converters conform to the
above rules.
 2004-12-13 02:32:46 +00:00
eg3p
a39d5c2ba3 changed all occurrences of 'Color.BLACK' to 'Color.black', since the former causes a runtime error on Mac OS X (for Java 1.3.1 at least). not sure why this is, but may be related to the bug mentioned at http://www.oxygenxml.com/forum/viewtopic.php?p=239 2004-08-19 14:59:06 +00:00
dchandler
e101cc8294 Now the Sambhota keyboard crashing bug. 
Fixed crashing bug reported by Teresa Lam.  Added tests so that I'm fairly
certain that no more crashing bugs exist.  Removed a marker for iffy code
after understanding that code via test cases.
 2004-07-05 04:36:35 +00:00
dchandler
6cbea9f894 Fixed crashing bug reported by Teresa Lam. Added tests so that I'm fairly 
certain that no more crashing bugs exist.  Removed a marker for iffy code
after understanding that code via test cases.
 2004-07-05 04:10:38 +00:00
dchandler
bff0e6b2fc Fixed TMW->Wylie/ACIP when multiple font sizes are in use. I was not 
incrementing the offset at which I was inserting text properly.
 2004-06-25 00:22:10 +00:00
dchandler
14fb449f95 I thought my earlier commit preserved font size info for TMW->ACIP/Wylie 
conversions.  It was only at a very coarse level.  The feature is now truly
here.
 2004-06-20 02:57:28 +00:00
dchandler
e18a4417dc Added a FIXME comment. 2004-06-12 02:26:28 +00:00
dchandler
9f78cabb18 TMW->{Wylie,ACIP} conversions now preserve font size information. 2004-06-12 02:09:28 +00:00
dchandler
7acbce3361 Added errors 142 and 143, which are produced when converting yig chung 
to a Unicode text file, which cannot support font size changes.
 2004-06-06 21:59:16 +00:00
dchandler
df262aa148 It is now a compile-time option whether to treat []- and {}-bracketed sequences 
as text to be passed through (without the brackets in the case of {}) literally,
which is the case by default because Robert Chilton requested it, or the old,
ad-hoc mechanism which could be useful for finding some ugly input.

Made a couple of error messages a little more verbose now that we have
short-message mode.
 2004-06-06 21:39:06 +00:00
dchandler
8a9271a3d8 I broke warning 507 into two warnings, one high-priority (512) and one 
low-priority (507).
 2004-05-01 20:49:53 +00:00
dchandler
31bdd39fec The TMW for 'da'i was converting to 'aad'i. Andres found this; it is bug 
945744.  I've made it more correct -- 'ad'i is now produced.  The wrong stack
is thought to be the root stack still.
 2004-05-01 19:11:15 +00:00
dchandler
1a055f3472 I don't think warning level "None" was really doing the trick. Fixed that. 
You can now customize the severities of all warnings, even 504 and 510.

When warning level is "None", scanning, i.e. lexical analysis, is faster.
 2004-04-25 00:37:57 +00:00
dchandler
e2d42f36eb Robert Chilton's experience inspired me to make the handling of errors and 
warnings in ACIP->Tibetan conversion much more configurable.  You can
now choose from short or long error messages, for one thing.  You can change
the severity of almost all warnings.  Each error and warning has an error code.
Errors and warnings are better tested.

The converter GUI has a new checkbox for short messages; the converter
CLI has a new mandatory option for short messages.

I also fixed a bug whereby certain errors were not being appended to the
'errors' StringBuffer.
 2004-04-24 17:49:16 +00:00
dchandler
cc5d096918 David Chapman's latest fix to tibwn.ini (clearing up an issue that Than or I 
dropped the ball on) introduced two lines for 8,95.  This is a bad thing, so
I've taken out the second line.  I've also introduced a check in
TibetanMachineWeb.java such that we'll know that tibwn.ini has no such
error in the future just by running 'ant clean jskad-run' and making sure that
the GUI is indeed visible.

I also updated the test baselines now that F03A and 0F82 are squared away.
 2004-04-24 13:23:56 +00:00
a1tsal
9e071ea178 Differentiated 0F82 (~M`) and F03A (nyi.zla editor's mark). 2004-04-21 10:04:11 +00:00
dchandler
0ee90a0fb0 Added many ACIP->TMW->ACIP tests. They found no bugs. 2004-04-17 17:28:26 +00:00
dchandler
63438d243b getACIP was getting EWTS, not ACIP. 2004-04-17 15:49:40 +00:00
dchandler
de3a19761e Fixes for javadoc tool. 2004-04-17 15:48:50 +00:00
dchandler
adcf9de952 Two new tests. 2004-04-17 15:14:46 +00:00
dchandler
1bfd3772e6 TMW->ACIP is much improved. V and W were confused, # and * were 
confused; many glyphs that should have yielded errors were not.

I've added a test case that transforms every TMW glyph save the one with
no TM mapping to ACIP.  I hand-checked that it was correct.

ACIP->TMW is fixed for # and *.  I never noticed it, but each needed an
extra swoosh (U+0F05).

Round-tripping would be good, as would testing real-world use of
TMW->ACIP.
 2004-04-14 05:44:51 +00:00
dchandler
56a02ba41d Fixed the worst TMW->ACIP bug, the one regarding U+0F04 and U+0F05. 
TMW->EWTS requires no context information, but TMW->ACIP does.
 2004-04-10 18:26:57 +00:00
dchandler
7eca276a62 TMW->Unicode conversions have changed; now using U+0F6A for the stacks 
whose EWTS transliteration begins with "R+".

ACIP->* conversions and test baselines were updated to deal with the
"r+..."=>"R+..."  change.
 2004-04-10 16:03:25 +00:00
dchandler
aff34174ab The new EWTS rule regarding R, W, and Y requires that these change. It 
may also require changes to the following, but I'm going to ask if it really
should or not.

// Y+Y~185,3~~6,98~1,109~6,120~1,123~1,125~6,106~6,113~f61,fbb
// Y+r~186,3~~6,99~1,109~6,120~1,123~1,125~6,106~6,113~f61,fb2
// Y+w~187,3~~6,100~1,109~6,120~1,123~1,125~6,106~6,113~f61,fad
// Y+s~188,3~~6,101~1,109~6,120~1,123~1,125~6,106~6,113~f61,fb6

// W+y~69,4~~7,79~1,109~8,121~1,123~1,125~8,107~8,114~f5d,fb1
// W+r~70,4~~7,80~1,109~8,121~1,123~1,125~8,107~8,114~f5d,fb2
// W+n~195,4~~7,81~1,109~8,120~1,123~1,125~8,106~8,113~f5d,fa3
// W+W~194,4~~7,82~1,109~8,120~1,123~1,125~8,106~8,113~f5d,fba
 2004-04-08 02:55:59 +00:00
dchandler
76356f4009 ACIP->Tibetan now gives an error when {?} is seen alone (not in {[?]} or {[*FOO?]}, but alone). Bug 860192 is fixed. 2004-03-15 00:49:01 +00:00
dchandler
542fb50bf1 The ~M and ~M` EWTS change had not fully been made. Someone submitted a bug report 911472 that alerted me to this. 2004-03-07 17:02:35 +00:00
dchandler
d436a4d462 Removed David Chapman's recently added line for U+0F82 -- a line for U+0F82 already existed, and the new line had incorrect TM and incorrect TMW mappings. I changed the existing line for U+0F82 to use the EWTS {~M`}. 2004-03-02 04:29:41 +00:00
a1tsal
8eaaeaa202 Fix careless error: I had the same TMW character for ~M and ~M`! 2004-02-22 09:14:56 +00:00
a1tsal
b14833b5b9 Change ^M to ~M to conform to spec. 
Introduce ~M` (for 0F82).
 2004-02-20 15:07:49 +00:00
dchandler
274e1736be Deleted cut-and-paste goof. 2004-01-17 19:45:31 +00:00
dchandler
c69ba26c60 TString now has tracks what Roman transliteration system it is using. Next up is to make ACIPConverter handle EWTS or ACIP TStrings. 2004-01-17 19:28:54 +00:00
dchandler
48b4c5cb07 Added a Unicode->ASCII dump for debugging *->Unicode conversions. To use it, use 'java -cp Jskad.jar org.thdl.util.VerboseUnicodeDump'. 2004-01-17 17:10:12 +00:00
dchandler
4dd40809a5 A user reported that q` caused a crash with TCC keyboard #1. Fixed. TCC keyboard #1 does not support q~ though. 2003-12-21 06:27:36 +00:00
dchandler
c1aa81e943 RFE 860190: ACIP->Unicode now gives a warning when it outputs something that can't be represented in TMW. 2003-12-16 07:45:40 +00:00
dchandler
848349fd3a More tests. 2003-12-15 08:16:06 +00:00
dchandler
e7a9e7968f ACIP->Unicode now uses two characters for consonants instead of one. This matches the dislike for characters like U+0F77 etc. 
ACIP->Tibetan was not giving an error for BCWA because it parsed like BCVA.  Fixed.
 2003-12-15 07:32:14 +00:00
dchandler
e9f7b2dfed If you want curly brackets around folio markers, you'll have to set 
the system property
thdl.acip.to.x.output.curly.brackets.around.folio.markers to true.
 2003-12-14 08:47:03 +00:00
dchandler
8664571577 Warnings were not being detected correctly. Fixed. 
ACIP->Unicode uses U+0020, ' ', for whitespace.  ACIP->TMW uses the
TMW whitespace for whitespace.
 2003-12-14 08:38:10 +00:00
dchandler
01e65176d4 Using less memory and time to figure out if warnings occurred. 2003-12-14 07:41:15 +00:00
dchandler
76c2e969ac Fixed ACIP->Unicode bug for YYE etc., things with full-formed 
subjoined consonants and vowels.

Fixed ACIP->TMW for YYA etc., things with full-formed subjoined
consonants.
 2003-12-14 07:36:21 +00:00
dchandler
f625c937ee ACIP {B} was not being treated like {BA}; instead, an error was resulting. All the five prefixes were affected. 2003-12-14 05:54:07 +00:00
dchandler
581643cf59 {DAN,\nLHAG} used to be treated like {DAN, LHAG} but that got broken. Fixed. 
Added tests for lexer's handling of ACIP spaces etc.
 2003-12-10 06:55:16 +00:00
dchandler
8e673bbc2c {NGA,} becomes {NGA\u0f0c,} now instead of {NGA\u0f0b,}. 
Note: ACIP->Unicode for {NGA,} was not giving the Unicode that {NGA\u0f0b,} gives before.
 2003-12-10 06:50:14 +00:00
dchandler
a466bad939 ACIP->TMW now supports EWTS PUA {\uF021}-style escapes. Our extended ACIP is thus TMW-complete and useful for testing. 2003-12-08 07:51:45 +00:00
dchandler
a39c5c12b0 ACIP->TMW now supports EWTS PUA {\uF021}-style escapes. Our extended ACIP is thus TMW-complete and useful for testing. 2003-12-08 07:15:27 +00:00
dchandler
b617f761d5 ACIP->TMW for {^GONG SA } used to fail; fixed. 2003-12-07 20:05:41 +00:00
dchandler
115534e688 ACIP->TMW for {^GONG SA } used to fail because we had \u0F38 in the ToWylie section. Now it's in the <?Input:Numbers?> section because I didn't want to introduce a new section. If WylieWord has trouble due to this misuse of the 'numbers' category, we'll introduce a new category, 'other'. 
TMW->EWTS improved as a result -- {\u0F38.gonga sa } is produced now where {\u0F38agonga sa } was once produced.  Even the better version is imperfect; see bug 855877.
 2003-12-07 19:40:59 +00:00
dchandler
4adf87c401 Updated comments only. 2003-12-06 20:36:56 +00:00
dchandler
3f18623977 Added comments only. 2003-12-06 20:26:45 +00:00
dchandler
6232ee9170 Added comments referring to a user guide in development now. 2003-12-06 20:26:15 +00:00
dchandler
c43e9a446b Revamped some ACIP->Tibetan error messages. 2003-12-06 20:19:40 +00:00
dchandler
c9c771d1ee ACIP {&}, as in {KO&HAm,}, is supported. 2003-11-30 02:18:59 +00:00
dchandler
ac412c994b Now {Pm} is treated like {PAm}; {Pm:} is like {PAm:}; {P:} is like {PA:}. 2003-11-30 02:06:48 +00:00
dchandler
e7c4cc1874 Updated to be in sync with latest EWTS draft. 2003-11-29 22:59:39 +00:00
dchandler
ffd041e32c ACIP->TMW and ACIP->Unicode now allow for Unicode escapes like K\u0F84. This means that the lack of support for ACIP's backslash, '\\', is mitigated because you can turn ACIP {K\} into ACIP {K\u0F84}. 
Support for U+F021-U+F0FF, the PUA that the latest EWTS uses, is not provided.

Also, we've traded some speed for memory -- DuffCode now uses bytes, not ints.
 2003-11-29 22:57:12 +00:00
dchandler
dfaae4be93 ACIP->TMW and ACIP->Unicode now allow for Unicode escapes like K\u0F84. This means that the lack of support for ACIP's backslash, '\\', is mitigated because you can turn ACIP {K\} into ACIP {K\u0F84}. 
Support for U+F021-U+F0FF, the PUA that the latest EWTS uses, is not provided.
 2003-11-29 22:56:18 +00:00
dchandler
16bfeac641 These issues are non-issues; removing these comments. 2003-11-25 00:31:33 +00:00
dchandler
d3d0ff23a8 Chris Fynn and Tony Duff answered my questions about U+0F3F and U+0F3E. 2003-11-25 00:28:18 +00:00
dchandler
b8608797aa Updated the code I used for testing to generate the file containing all glyphs in TM and all glyphs but one in TMW. 2003-11-24 05:59:32 +00:00
dchandler
5d053b41fe Found another inconsistency between Unicode and the TM/TMW docs. I've sent e-mail to Tony Duff asking who's right, but I'm putting this in the errata under the assumption that even if Unicode is wrong, Unicode's wrong view will somehow rule the day. 
Also, TMW->EWTS now generates \uF021-\uF0FF or \u0F00-\u0FFF escapes when appropriate.  A few TMW glyphs still give errors.

Also, there's now a test to be sure that TM<->TMW and TMW->EWTS won't break in the future (except for the one glyph in TMW that isn't in TM, that one isn't tested).  The baselines have not been hand-verified, but changes will be detected.
 2003-11-24 05:49:15 +00:00
dchandler
9a247f5932 N+D+Ya, not N+D+ya, w+Wa, not w+wa .. use W, R, and Y where appropriate. 2003-11-24 04:55:11 +00:00
dchandler
1ec668c018 Dza is not in the latest EWTS draft. 2003-11-24 04:28:55 +00:00
dchandler
f76c089366 Using Y, R, and W everywhere needed. R+... is never needed in TM/TMW, I concluded (with 50% certainty). 2003-11-24 04:05:59 +00:00
dchandler
08c676c186 Bug fixes. Plus, now 99% in sync with the new EWTS draft. Search for 'DLC' to find a few open issues. 
Readded the line for reversed dza; it should never have been deleted, as that breaks TM<->TMW.  I tested the whole mapping by hand once; this incident shows that automation is very helpful.

'{' and '}' were swapped...

The Unicode for something was "", not "none".

+R, +W, +Y, R+ now in use (though more testing is needed)
 2003-11-24 02:40:40 +00:00
dchandler
216c5b0d54 Fixed TWM->Wylie for achen. I even tested this by pretending achen could take a da prefix (when in reality it takes no prefixes). 2003-11-23 01:22:27 +00:00
dchandler
8d4fb5d13f We crashed before when '~' was entered. 2003-11-14 04:50:55 +00:00
dchandler
b59b86fd73 Commented this to mention some recent testing. 2003-11-11 03:45:58 +00:00
dchandler
4023be9612 Better prettyprinting. Untested. 2003-11-11 03:43:26 +00:00
dchandler
4e6a9c299f ACIP % {MTHAR%} and o {Ko} and ^ {^GONG SA} are now supported. A % always causes a warning. 2003-11-11 03:43:11 +00:00
dchandler
2cb90bd231 ACIP->Tibetan converters now warn every time {%} is encountered that U+0F14 might've been intended. 
The Unicode for ACIP {o} is U+0F37.
 2003-11-09 23:15:58 +00:00
dchandler
04816acb74 ACIP->Unicode was broken for KshR, ndRY, ndY, YY, and RY -- those 
stacks that use full-form subjoined RA and YA consonants.

ACIP {RVA} was converting to the wrong things.

The TMW for {RVA} was converting to the wrong ACIP.

Checked all the 'DLC' tags in the ttt (ACIP->Tibetan) package.
 2003-11-09 01:07:45 +00:00
dchandler
8193cef5d1 Better comments. 2003-11-09 01:07:07 +00:00
dchandler
3fa417d3ee phywI, phywU, drwI and drwU now produce vowels and subjoined a-chungs. The Tibetan! 5.1 docs say I and U are not applicable to these stacks, but I say Jskad lets the user decide what's applicable. If you disagree, be sure to give an error message before dropping the I or U request -- we were silent. 2003-11-08 21:53:34 +00:00
dchandler
e058d6252e phywu and drwu now produce zhabs-kyus. The Tibetan! 5.1 docs say the zhabs-kyu is not applicable to these stacks, but I say Jskad lets the user decide what's applicable. If you disagree, be sure to give an error message before dropping the zhabs-kyu request -- we were silent. 2003-11-08 21:48:08 +00:00
dchandler
55aaeef9d0 l+h+wu now produces a zhabs-kyu. The Tibetan! 5.1 docs say the zhabs-kyu is not applicable to l+h+w, but I say Jskad lets the user decide what's applicable. If you disagree, be sure to give an error message before dropping the zhabs-kyu request -- we were silent. 2003-11-08 21:23:50 +00:00
dchandler
06edf17b04 Once again, the wrong 'dreng-bu glyphs were listed in the Tibetan! 5.1 docs -- they were na-ro glyphs, actually. 2003-11-08 21:17:18 +00:00
dchandler
74d6bc61ab The wrong 'dreng-bu glyphs were listed in the Tibetan! 5.1 docs -- they were na-ro glyphs, actually. 2003-11-08 20:25:16 +00:00
dchandler
a0ae0bf70d Fixes bug 800164. Jskad users can now enter t+r+n on the keyboard. Wylie Word should work for t+r+n too. 2003-11-08 17:50:10 +00:00
dchandler
e3f1ed5914 Removed a DOS EOF character (^Z). I haven't a clue how it crept in -- the lexer doesn't let that kind of thing get into tsheg bars. 2003-10-27 13:58:45 +00:00
dchandler
94a43d3f39 Now anything not clearly native Tibetan is colored green when coloring is enabled. G'EEm is "native", though -- the only "vowel" that implies non-nativeness is {:}, as in {KA:}. 2003-10-26 18:56:48 +00:00
dchandler
5c36dd81d3 Fixed bug 830332, "Convert selected ACIP=>Tibetan busted". 2003-10-26 18:25:25 +00:00
dchandler
e74547d743 GA-YOGS now parses like G-YOGS and GAYOGS do. 2003-10-26 18:06:38 +00:00
dchandler
61cf19932e ACIP {B5} and {7'} were problematic; that's fixed. 2003-10-26 17:47:35 +00:00
dchandler
ad7b20e485 Added yet more metadata. 2003-10-26 16:05:30 +00:00
dchandler
1550fee41a Removed garbage. 2003-10-26 16:05:07 +00:00
dchandler
fe33d67573 Added more metadata. There are 35 million+ tsheg bars here. 2003-10-26 15:35:08 +00:00
dchandler
050666d735 I'm committing this at 1:55 am EST on Sunday, October 26, 2003. There 
is no compelling technical reason, but this way I get to have two
commits that are both before and after each other.

Freaky.
 2003-10-26 06:56:12 +00:00
dchandler
31b3020d07 Added a test case that runs almost all the tsheg bars from all 
non-reference, publicly available ACIP files (hundreds of megabytes of
them) through the converter.  The frequencies of these tsheg bars in
in the file, too.
 2003-10-26 06:02:48 +00:00
dchandler
7ba1ad0735 Added a mechanism for end users to have the ACIP/EWTS=>Tibetan converters print all tsheg bars or all unique tsheg bars to standard output. This will be useful for getting a list of all the tsheg bars in ACIP texts, e.g., which can then go into PackageTest.java. A lot of postprocessing would be required to get frequency counts, but you could do it with a perl script, awk, etc. 2003-10-26 02:42:06 +00:00
dchandler
ef24c608bf Added a mechanism for end users to customize ACIP/EWTS=>Tibetan conversions by giving a list of substitutions to be performed. E.g., when I invoke Jskad via 'java -Dorg.thdl.tib.text.ttt.VerboseReplacementMap=false -Dorg.thdl.tib.text.ttt.ReplacementMap="KAsh=>K+sh" -jar Jskad.jar', then the ACIP KAsh becomes K+sh automatically. 
This mechanism is for Andres (who noticed KAsh=>K+sh in practice) and power users only, and not power users until I document the thing outside of the source code.
 2003-10-26 02:17:19 +00:00
dchandler
6bda550157 The ACIP "BNA" was converting to B-NA instead of B+NA, even though NA cannot take a BA prefix. This was because BNA was interpreted as root-suffix. In ACIP, BN is surely B+N unless N takes a B prefix, so root-suffix is out of the question. 
Now Jskad has two "Convert selected ACIP to Tibetan" conversions, one with and one without warnings, built in to Jskad proper (not the converter, that is).
 2003-10-26 00:32:55 +00:00
dchandler
d99ae50d8a The ACIP "BNA" was converting to B-NA instead of B+NA, even though NA cannot take a BA prefix. This was because BNA was interpreted as root-suffix. In ACIP, BN is surely B+N unless N takes a B prefix, so root-suffix is out of the question. 
Now Jskad has two "Convert selected ACIP to Tibetan" conversions, one with and one without warnings, built in to Jskad proper (not the converter, that is).
 2003-10-26 00:24:28 +00:00
dchandler
306cf2817c Private correspondence with Robert Chilton led to me to add and remove a few prefix rules. BLC and BGL are here, BLK, BLG, BLNG, BLJ, BNG, BJ, BNY, BN, and BDZ are gone. 
Added a few new tests.
 2003-10-25 21:47:34 +00:00
dchandler
f106deb884 Private correspondence with Robert Chilton led to me to add and remove a few prefix rules. BLC and BGL are here, BLK, BLG, BLNG, BLJ, BNG, BJ, BNY, BN, and BDZ are gone. 
Added a few new tests.
 2003-10-25 21:40:21 +00:00