about summary refs log tree commit diff stats
Commit message (Collapse)AuthorAgeFilesLines
* Fix crash in openEditor if SIGINT was deliveredbptato2023-06-265-3/+29
| | | | | | | | | | | nvi for example sets ISIG and traps SIGINT. Without this patch, this would propagate to all processes in the same process group and kill them. (It still does, but we set a signalHandler to ignore that.) Still not perfect, because for some reason we receive WIFSIGNALED even if the signal did not actually kill the editor. For now, we just treat this as a success.
* hacks for readline so that it at least kind of worksbptato2023-06-253-4/+22
| | | | Proper fix needed later
* rendertext: do not rewind if no other charsets are availablebptato2023-06-251-1/+1
| | | | Fixes crash when the decoder fails to decode all given charsets.
* css/values: remove stray eprintbptato2023-06-251-2/+0
|
* twtstr: remove unused functionbptato2023-06-251-13/+0
|
* cssparser: remove SyntaxErrorbptato2023-06-252-24/+34
| | | | return DOMException instead
* buffer: only load images if config allows itbptato2023-06-251-1/+4
|
* png: fix 4 bit grayscale & <8 bit indexed colorbptato2023-06-251-6/+17
|
* css/values: use Result instead of exceptionsbptato2023-06-253-312/+344
|
* cookie: return error on invalid domainbptato2023-06-252-6/+7
|
* Improve broken cookie handling, add multipart to formbptato2023-06-255-17/+126
| | | | | | | | Now it's not as horribly broken as before (but it's still far from perfect). We can at least log in to sr.ht (hooray). The form multipart part is straightforward, just pass what we used to pass long ago before I broke multipart.
* pager: alert on invalid regexbptato2023-06-241-12/+13
|
* Option -> Opt in config, return error in compileSearchRegexbptato2023-06-245-55/+63
|
* Only set terminal title when it is a ttybptato2023-06-241-2/+5
| | | | Avoid writing the escape sequence when dumping files.
* Rename readme.md to upper casebptato2023-06-241-0/+0
| | | | So that sourcehut can find it
* Public domainbptato2023-06-242-0/+28
|
* Fix potential crash on startupbptato2023-06-241-4/+7
|
* Working Nim-QuickJS GC integrationbptato2023-06-246-44/+137
| | | | I believe this works correctly. Hopefully I'm not wrong.
* Makefile: remove random echo statementbptato2023-06-241-1/+0
|
* Add dir to qjs target, rename makefilebptato2023-06-231-1/+1
|
* radixtree: optimize searchingbptato2023-06-232-26/+54
|
* eprint: simplifybptato2023-06-231-33/+23
|
* term: fix clearEnd regressionbptato2023-06-231-1/+3
| | | | Sending EL at EOL clears the last character, so we avoid doing that.
* Fix module loading bugsbptato2023-06-223-8/+23
|
* Include QuickJS sources in the repositorybptato2023-06-2220-30/+74831
| | | | Taken from txiki.js, so it includes zamofex's top-level await patch.
* Remove LegacyJSError, use enums for storing symbolsbptato2023-06-221-57/+45
|
* Replace nim_finalize_for_js with collectInteropCyclesbptato2023-06-201-51/+23
| | | | | | Pros: maybe doesn't corrupt memory Cons: leaks memory instead Maybe we should just use destructors...
* Do not GC_ref JS object opaquesbptato2023-06-201-1/+0
| | | | So that the finalizer is actually triggered
* makefile: separate objdir for all targetsbptato2023-06-201-2/+4
|
* Do not allow text() twice, etc...bptato2023-06-205-23/+50
|
* Do not pass JSContext to getJSValuebptato2023-06-201-5/+5
|
* Fix fromJS Opt bugsbptato2023-06-191-9/+10
|
* Add Response.ok, url, fix type confusion with statusbptato2023-06-193-7/+17
|
* Reject fetch promise on network errorbptato2023-06-197-27/+33
| | | | Instead of setting the non-standard res variable.
* Get rid of the .jserr pragmabptato2023-06-1912-277/+248
|
* Fix memory leak with alt pointersbptato2023-06-191-1/+6
| | | | Must unref them if unused.
* Rework JS exception systembptato2023-06-1916-178/+480
| | | | | Now we use Result for passing exceptions to JS. As a result, we can finally get rid of the .jserr pragma.
* input size as presentational hintbptato2023-06-182-0/+12
|
* Use utils/opt in toml parserbptato2023-06-185-165/+207
|
* Add typeerror to jsgctorbptato2023-06-181-1/+1
|
* Revert "renderdocument: bgcolor blending"bptato2023-06-152-26/+21
| | | | | This reverts commit c13a4cf4e144f7ab36167ebbb1333f7faeed11b3. Clearly it's broken. Needs a fix.
* Cleanups & bug fixes involving promisesbptato2023-06-1510-98/+164
| | | | | Now a promise returning nil doesn't just leave the rest of the then chain hanging. Hooray.
* renderdocument: bgcolor blendingbptato2023-06-132-21/+26
|
* Make Result.text, json return promisebptato2023-06-136-25/+66
|
* Fix various layout bugs, prepare for image supportbptato2023-06-1211-136/+212
|
* Add readBlob, fix WebFile serializationbptato2023-06-113-16/+39
| | | | It's still kind of a mess, but it kind of works?
* Fix -cha-center moving by negative xbptato2023-06-101-4/+2
| | | | | In other words, if the element is wider than it's container, we don't do anything at all.
* Fix color U, slightly change contrast algorithmbptato2023-06-103-10/+19
|
* Use width, height for canvas in cssbptato2023-06-101-1/+1
|
* approximateANSIColor: fix possible underflowbptato2023-06-102-3/+8
|
generated by cgit-pink 1.4.1-2-gfad0 (git 2.36.2.497.gbbea4dcf42) at 2025-01-09 21:26:14 +0000