aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* use iswspace()/iswpunct() to find word delimitersLauri Tirkkonen2019-03-153-5/+7
| | | | | | this inverts the configuration logic: you no longer provide a list of delimiters -- all space and punctuation characters are considered delimiters, unless listed in extrawordchars.
* replace utf8strchr with wcschrLauri Tirkkonen2019-03-153-22/+4
|
* be silent about explicitly unhandled mouse modesLauri Tirkkonen2019-03-131-0/+1
|
* simplify (greedy) font caching allocating a bitHiltjo Posthuma2019-03-031-7/+2
| | | | | | POSIX says: "If ptr is a null pointer, realloc() shall be equivalent to malloc() for the specified size."
* style: remove double empty newlinesHiltjo Posthuma2019-03-032-3/+0
|
* fix use after free in font caching algorithmmagras2019-03-031-6/+9
| | | | | | | | | | | | | | | | | | Current font caching algorithm contains a use after free error. A font removed from `frc` might be still listed in `wx.specbuf`. It will lead to a crash inside `XftDrawGlyphFontSpec()`. Steps to reproduce: $ st -f 'Misc Tamsyn:scalable=false' $ curl https://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-demo.txt Of course, result depends on fonts installed on a system and fontconfig. In my case, I'm getting consistent segfaults with different fonts. I replaced a fixed array with a simple unbounded buffer with a constant growth rate. Cache starts with a capacity of 0, gets increments by 16, and never shrinks. On my machine after `cat UTF-8-demo.txt` buffer reaches a capacity of 192. During casual use capacity stays at 0.
* better Input Method Editor (IME) supportIvan Tham2019-02-123-17/+54
| | | | | | | | | | | | Features: - Allow input methods swap with hotkey (E.g. left ctrl + left shift). - Over-the-spot pre-editing style, pre-edit data placed over insertion point. - Restart IME without segmentation fault. TODO: - Automatically pickup IME if st started before IME
* bump version to 0.8.2Hiltjo Posthuma2019-02-091-1/+1
|
* config: add Shift+Insert as selpaste() againHiltjo Posthuma2019-01-101-0/+1
| | | | | This was changed before in: commit 20f713548de451b67db3306cf8cf7b2f38fee05c on Wed Jan 25 19:17:38 2017
* Let the user specify CPPFLAGSParide Legovini2019-01-101-2/+2
| | | | | | | | | | | This complements the work done in d4928ed, allowing the user to specify the preprocessor flags with the CPPFLAGS environment variable. This is useful for example to specify preprocessor macros with -D. CFLAGS could be used instead, but CPPFLAGS is more correct and is expected to be honored in some cases. For example, the helper scripts to build Debian packages make use of CPPFLAGS, but the variable is currently being ignored unless manually appended to CFLAGS.
* Set the path of pkg-config in a variable instead of hardcoding itParide Legovini2019-01-041-4/+6
| | | | | In this way the path of pkg-config can be overridden from the command line. This is useful for example when cross-compiling.
* Makefile: fix dependencies on config.hHiltjo Posthuma2019-01-041-1/+1
| | | | patch by Younes Khoudli (changed slightly). Thanks
* output child WEXITSTATUS/WTERMSIG on abnormal terminationLauri Tirkkonen2018-12-111-2/+4
|
* fix memory leak in xloadcols()Hiltjo Posthuma2018-11-041-3/+3
| | | | | | reported by Avi Halachmi (:avih)" <avihpit@yahoo.com> patch slightly changed by me.
* st: small typofix in commentHiltjo Posthuma2018-11-041-1/+1
|
* small code-style fixHiltjo Posthuma2018-09-111-1/+2
|
* Remove the ISO 14755 featureQuentin Rameau2018-09-114-32/+0
| | | | | | And move it to the patches section. Keeping it would force to add an exec pledge on OpenBSD, and some people think it's bloated, so bye!
* Revert "Simplify cursor color handling"Hiltjo Posthuma2018-07-171-12/+18
| | | | This reverts commit 1911c9274d9b03f3d7999c6ce26e2d5169642d26.
* Revert "Make cursor follow text color"Hiltjo Posthuma2018-07-172-7/+7
| | | | This reverts commit b51bcd5553af3db394014efbd78acf7828fa48ff.
* Revert "Fix crash when cursor color is truecolor"Hiltjo Posthuma2018-07-171-13/+4
| | | | This reverts commit 5535c1f04c665c05faff2a65d5558246b7748d49.
* Fix crash when cursor color is truecolorJules Maselbas2018-07-151-4/+13
| | | | Reported-by: Ivan Tham <pickfire@riseup.net>
* Make cursor follow text colorJules Maselbas2018-07-142-7/+7
|
* Simplify cursor color handlingJules Maselbas2018-07-141-18/+12
|
* Fix crash on resizeJules Maselbas2018-06-301-4/+8
| | | | | Prevent to realloc xw.specbuc with a negative number of col. Add proper hints for the minimal size, for one character.
* config.mk: remove extra newline before EOFHiltjo Posthuma2018-06-021-1/+0
|
* code-style for pledge(2)Hiltjo Posthuma2018-05-251-4/+4
| | | | feedback from Klemens, thanks
* Pledge on OpenBSDHiltjo Posthuma2018-05-252-0/+14
|
* error message style and use strerror in a few placesHiltjo Posthuma2018-03-292-24/+24
|
* st -v: remove years and copyright textHiltjo Posthuma2018-03-291-1/+1
|
* set sel.alt in selstart instead of selextendDaniel Tameling2018-03-291-1/+1
|
* bump version to 0.8.1Hiltjo Posthuma2018-03-201-1/+1
|
* fix regression by selecting clipboard textHiltjo Posthuma2018-03-201-0/+3
| | | | | | | "restore the old behaviour that the primary doesn't get deleted by a simple left click" Patch by Daniel Tameling <tamelingdaniel@gmail.com>, thanks!
* don't modify argv, use a counterHiltjo Posthuma2018-03-201-10/+12
| | | | | on some platforms (OpenBSD) this changes the exposed argv in tools using the kvm_* interface, such as ps and pgrep.
* selextend: clarify: !sel.mode == SEL_IDLEHiltjo Posthuma2018-03-171-1/+1
|
* clipcopy: no need to check for free(NULL), set to NULL after freeHiltjo Posthuma2018-03-171-2/+2
|
* minor code-style: whitespace fixesHiltjo Posthuma2018-03-161-2/+1
|
* Fix title initializationQuentin Rameau2018-03-161-5/+5
|
* Fix regression from 69e32a6 when setting title.Quentin Rameau2018-03-161-1/+1
|
* LICENSE: fix a few yearsHiltjo Posthuma2018-03-141-3/+3
|
* update LICENSE: major contributorsHiltjo Posthuma2018-03-141-1/+4
|
* Makefile: add all files to make distHiltjo Posthuma2018-03-141-1/+3
|
* bump version to 0.8Hiltjo Posthuma2018-03-141-1/+1
|
* use math.h for ceilfHiltjo Posthuma2018-03-091-1/+1
|
* xhints: no need to initialize sizehHiltjo Posthuma2018-03-091-1/+1
|
* regression: include termios.h for tcsendbreak etcHiltjo Posthuma2018-03-091-0/+1
|
* General cleanupDevin J. Pohly2018-02-252-26/+21
| | | | | | | Simplifies logic in a couple places and removes a redundant function call. Signed-off-by: Devin J. Pohly <djpohly@gmail.com>
* Clean up #includesDevin J. Pohly2018-02-253-9/+4
| | | | Signed-off-by: Devin J. Pohly <djpohly@gmail.com>
* Reduce visibility wherever possibleDevin J. Pohly2018-02-254-44/+43
| | | | | | | | When possible, declare functions/variables static and move struct definitions out of headers. In order to allow utf8decode to become internal, use codepoint for DECSCUSR extension directly. Signed-off-by: Devin J. Pohly <djpohly@gmail.com>
* Limit usage of extern to config.h globalsDevin J. Pohly2018-02-254-27/+29
| | | | | | | | Prefer passing arguments to declaring external global variables. The only remaining usage of extern is for config.h variables which are needed in st.c instead of x.c (where it is now included). Signed-off-by: Devin J. Pohly <djpohly@gmail.com>
* Remove x.c dependency on termDevin J. Pohly2018-02-253-39/+39
| | | | | | | | | | | The xinit function only needs to the rows/cols, so pass those in rather than accessing term directly. With a bit of arithmetic, we are able to avoid the need for term.row and term.col in x2col, y2row, and xdrawglyphfontspecs as well, completing the removal. Term is now fully internal to st.c. Signed-off-by: Devin J. Pohly <djpohly@gmail.com>