summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Forgot to set a file position on characters being returned fromSimon Tatham2004-08-20
| | | | | | macro expansions. [originally from svn r4489]
* Remove a \#{FIXME} comment in \cfg{paper-chapter-top-space}; the checkinJacob Nevins2004-08-09
| | | | | | | of bk_paper.c 1.29 [r4330] appears to have fixed this. [originally from svn r4432] [r4330 == 8aac8532019ac6cee701d48ceb35473280219a0e]
* `make release' tweak to prevent tarball containing trailing NULs (fix fromJacob Nevins2004-08-09
| | | | | | putty/mkunxarc.sh) - apparently this upsets some older gzips (1.2.4). [originally from svn r4431]
* Explicitly bless in the documentation, and fix in the man backend,Simon Tatham2004-08-06
| | | | | | | the practice of using \dt and \dd in anything other than the obvious interleaving. [originally from svn r4417]
* Make sure (paragraph)->private_data is defined in all cases, on generalJacob Nevins2004-08-06
| | | | | | | principles (it may be unnecessary). Perhaps we should do this for (word)->private_data too. [originally from svn r4412]
* Fix crash in Info backend: if a \[Kk] to a \B or \n was followed in aJacob Nevins2004-08-06
| | | | | | paragraph by another \[Kk], Bad Things would happen. [originally from svn r4411]
* dup_word_list() does the right thing if passed NULL (on general principles)Jacob Nevins2004-08-06
| | | | [originally from svn r4410]
* Fix for an uninitialised structure member spotted by valgrind (probably didn'tJacob Nevins2004-08-04
| | | | | | have any real effect). [originally from svn r4401]
* Explicit bounds checking on the bmp[] array.Simon Tatham2004-08-04
| | | | [originally from svn r4400]
* Add an explicit cast to prevent a compiler warning.Simon Tatham2004-08-04
| | | | [originally from svn r4399]
* Colin spotted a sizeof/lenof confusion which we think has beenSimon Tatham2004-08-04
| | | | | | causing segfaults on strange architectures. [originally from svn r4398]
* More careful context hash calculation which doesn't depend on sizeSimon Tatham2004-07-30
| | | | | | of `unsigned long'. Should help Debian build problems (bug#259504). [originally from svn r4377]
* Division by zero (which criminally failed to give rise to any kindSimon Tatham2004-07-17
| | | | | | | of signal!) was causing massive misplacement of the single line of text on a one-line page in PS and PDF. [originally from svn r4330]
* Cut-and-paste error was preventing PS and PDF filename configSimon Tatham2004-07-17
| | | | | | options from being processed in some circumstances. [originally from svn r4329]
* Tweak for \cq (probably unnecessary)Jacob Nevins2004-07-16
| | | | [originally from svn r4328]
* Right; I'm finally sick of typing \q{\cw{foo}}, so I've invented aSimon Tatham2004-07-16
| | | | | | shorthand command \cq{foo}. [originally from svn r4327]
* Failed to initialise state.curr_topic early enough, causingSimon Tatham2004-07-13
| | | | | | | uninitialised-memory references and a segfault. Quite why this never happened when I tested it at home, I have no idea! [originally from svn r4326]
* Apparently (according to a user) WinZip complains of `trailingSimon Tatham2004-07-05
| | | | | | | | | garbage' when handling a GNU tar archive. Using the -o flag when building the tar file creates an archive which the user said WinZip had no trouble with, and GNU tar seems just as happy with that, so `make release' now uses -o. [originally from svn r4323]
* Avoid generating an index section if there actually is no index.Simon Tatham2004-07-02
| | | | [originally from svn r4321]
* Sort out error handling everywhere a charset name is converted intoSimon Tatham2004-06-27
| | | | | | an integer charset ID. [originally from svn r4317]
* Free up all the data we allocated during the HTML backend.Simon Tatham2004-06-27
| | | | [originally from svn r4316]
* Support non-breaking spaces.Simon Tatham2004-06-27
| | | | [originally from svn r4315]
* Config directives for all the extra configurable things in the newSimon Tatham2004-06-27
| | | | | | HTML backend. [originally from svn r4314]
* Fix up the restrict_charset and output_charset configuration for theSimon Tatham2004-06-27
| | | | | | | | | | | | | | HTML backend. After an argument with James in which it transpired that we each thought it was absolutely obvious which one \cfg{html-charset} ought to map to, and each of us was thinking of a different one, I've decided that this implies it _isn't_ obvious, and hence there is no single \cfg{html-charset} directive at all. Instead, we have \cfg{html-restrict-charset} and \cfg{html-output- charset}, and any naive user who `just wants to change character set' is just going to have to RTFM and figure out which one they mean. [originally from svn r4313]
* nit: s/case-sensitive/case-insensitive/ in comment about nasty indexingJacob Nevins2004-06-21
| | | | | | hack. [originally from svn r4311]
* Various HTML backend cleanups.Simon Tatham2004-06-20
| | | | [originally from svn r4308]
* Introduce a configurable option to select the HTML flavour. AlsoSimon Tatham2004-06-20
| | | | | | | fiddle with various small aspects of the output so that it actually validates in all supported flavours. [originally from svn r4307]
* Improve the naming of fragment IDs for numbered list elements andSimon Tatham2004-06-20
| | | | | | bibliography entries. [originally from svn r4305]
* Minimal documentation of all the new paper backend \cfg{} directives.Jacob Nevins2004-06-20
| | | | | | | | | Contains many FIXMEs, both for incomplete documentation and incomplete code. Couple of other minor tweaks elsewhere. [originally from svn r4304]
* We seem to have ended up with duplicate documentation of --man. Gone.Jacob Nevins2004-06-19
| | | | [originally from svn r4303]
* Ensure that every \cfg{} directive documented so far has appropriate \IMJacob Nevins2004-06-19
| | | | | | directives. [originally from svn r4302]
* Fairly wholesale reorganisation of the \cfg{} documentation for the text,Jacob Nevins2004-06-19
| | | | | | | | WinHelp, man, and info backends, and organisation into sections (now that we have rather more directives documented). I've not changed the actual text much if at all. [originally from svn r4301]
* In the text backend, multiple characters can now be used for underlining.Jacob Nevins2004-06-19
| | | | [originally from svn r4300]
* The Unicode OVERLINE in the text backend documentation was probably supposedJacob Nevins2004-06-19
| | | | | | to be quoted. [originally from svn r4299]
* Note that \cfg{quotes} and friends apply to the output in its entirety andJacob Nevins2004-06-19
| | | | | | can't be changed partway through. Also, add a cross-reference from \q{}. [originally from svn r4298]
* Note that no index is generated by the text backend.Jacob Nevins2004-06-19
| | | | [originally from svn r4297]
* Minor correction to --ps and --pdf descriptions.Jacob Nevins2004-06-19
| | | | [originally from svn r4296]
* Initial rough documentation of some of the newer \cfg{} directives, namely:Jacob Nevins2004-06-19
| | | | | | | | | | | | - \cfg{quotes} - text, info, man, and winhelp backends: - charset - quotes and bullets - various WinHelp miscellanea Could stand some rearrangement and expansion, plus I've yet to address the paper or HTML backends. [originally from svn r4295]
* Moderately nasty workaround for the fact that Windows Help's indexSimon Tatham2004-06-19
| | | | | | | | | mechanism is case-insensitive with respect to sorting the list of index entries. We now append nonbreaking spaces to the ends of some index terms to ensure they are considered distinct. (Nasty, but it seems to work.) [originally from svn r4294]
* Add a batch of assertions to ensure no more index terms fail to haveSimon Tatham2004-06-19
| | | | | | | | | their anchors generated. Index bugs of this type can be terribly subtle and go unnoticed for ages; I'm lucky Jacob spotted the first one of the type, but I'm now going to be careful to avoid any further ones. [originally from svn r4293]
* I had somehow managed to leave index entry anchors out of sectionSimon Tatham2004-06-19
| | | | | | titles. Ahem. [originally from svn r4292]
* Add a clash-checking mechanism to ensure we never generate the sameSimon Tatham2004-06-19
| | | | | | fragment name twice for any reason. [originally from svn r4291]
* Support \I\c, \I\cw. (Haven't bothered with \I\e since the emphasis may asJacob Nevins2004-06-19
| | | | | | well outweigh the "invisibility".) [originally from svn r4290]
* s/text-quotes/man-quotes/. Damn cut and paste. Thanks Jacob.Simon Tatham2004-06-16
| | | | [originally from svn r4288]
* The Halibut manual contained at least one instance of two indexSimon Tatham2004-06-13
| | | | | | | | | | | | | | terms (intentionally) differing only in case, which were being silently folded into one by the case-insensitive index tag comparison. Halibut now warns in this situation (but then folds them anyway, which I think is better than silently generating an index containing many case-distinct forms of the same word - I imagine it's very easy to do that by mistake). The manual has been fixed to explicitly define distinct keywords (in the case I spotted and in five other cases picked up by the new warning!), and also documents this issue and how to work with it. [originally from svn r4279]
* Er, don't forget to document the disallowing of \q within \c. Ahem.Simon Tatham2004-06-12
| | | | [originally from svn r4278]
* Enforce that \q may not be used anywhere within \c. It shouldn't beSimon Tatham2004-06-12
| | | | | | | | | | | | | necessary since the whole point of \c should be that the user wants to exercise exact control over the glyphs used, and forbidding it has the useful effect of relieving some backends of having to make difficult decisions: it means the text backend doesn't have to nest two pairs of identical quotes, and the paper backends don't have to downgrade their quote characters if (as is perfectly plausible) the fixed-pitch font doesn't support the same range as the body text fonts. [originally from svn r4277]
* Switch the memory allocation macros from the Halibut onesSimon Tatham2004-06-12
| | | | | | | | | | | | | (mknew/mknewa/resize) to the PuTTY ones (snew/snewn/sresize). snewn and mknewa have their arguments opposite ways round; this may make the change initially painful but in the long term will free me of a nasty context switch every time I move between codebases. Also sresize takes an explicit type operand which is used to cast the return value from realloc, thus enforcing that it must be correct, and arranging that if anyone tries to compile Halibut with a C++ compiler there should be a lot less pain. [originally from svn r4276]
* Initial checkin of the shiny new rewritten-from-scratch HTML backSimon Tatham2004-06-12
| | | | | | | | | | | | | | | | | | end. There's a lot more _potentiality_ for new features than there are actual new features just yet, but future highlights include: configurable flavour of HTML (3.2, 4, XHTML Transitional or Strict), proper character set support (this is half way there already), and more flexible allocation of sections between multiple HTML files. Meanwhile, immediate benefits include correct handling of special characters within `author' and `description' strings, omission of the filename part in hyperlinks within the same HTML file (in particular, this means a single output file is now totally independent of its filename), and hyperlinks to the index from the top-level contents page (I'm amazed nobody has complained at the lack of this yet!). There are no doubt some shiny new bugs as well, but I'll never find them unless people start using the thing... [originally from svn r4275]
* Preferred MIME name for ASCII is "US-ASCII", not "ANSI_X3.4-1968". Oops.Simon Tatham2004-06-03
| | | | | [originally from svn r4269] [this svn revision also touched charset,filter,timber]
generated by cgit v1.1 at 2026-05-23 13:19:59 +0000