Browse source

Changelog

Query
From:
To:
Branch:
Path:
Message:
BugID:
User: +
Format:
: 1 checkin (+4/-3) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 4 checkins (+35/-2) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 4 checkins (+11/-4) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+3/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 71 checkins (+262/-116) by 2 people : 1 checkin (+1/-13) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 2 checkins (+17/-4) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 5 checkins (+6/-1) by 2 people : 1 checkin (+5/-4) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+15/-7) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+4/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+2/-1) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 8 checkins (+35/-6) by 2 people : 8 checkins (+25/-5) by 2 people : 7 checkins (+24/-5) by 2 people : 1 checkin (+27/-2) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 5 checkins (+232/-28) by 2 people : 1 checkin (+5/-2) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 2 checkins (+9/-4) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+7/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 15 checkins (+125/-99) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 8 checkins (+116/-93) by 2 people : 1 checkin (+2/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+7/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+6/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 7 checkins (+30/-25) by 3 people : 1 checkin (+2/-1) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 2 checkins (+47/-2) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+20/-15) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+77/-49) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+3/-0) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+14/-6) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 1 checkin (+2/-1) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 3 checkins (+87/-81) by 2 people : 6 checkins (+93/-67) by 2 people : 7 checkins (+58/-39) by 3 people : 2 checkins (+13/-1) by Henrik Grubbström (Grubba) <grubba@grubba.org> : 27 checkins (+1047/-950) by 2 people : 5 checkins (+217/-138) by 2 people : 1 checkin (+3/-1) by Henrik Grubbström (Grubba) <grubba@grubba.org> Lines added/removed recently

Yesterday; Sunday 16 June 2019

2019-06-16 14:48:50 (22 hours ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Stdio.Buffer: Fixed input_from() with non-Stdio.File objects.

The return value from input_from() was always -1 when it was called
with non-Stdio.File objects (even if it successfully read data).

This issue is triggered by the testsuite for the Filesystem module.

Saturday 15 June 2019

2019-06-15 11:39:30 (2 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

AutoDoc: Support using stand-alone doc for nested classes too.

2019-06-15 04:25:00 (2 days ago) by Stephen R. van den Berg <srb@cuci.nl>

Gz.Pipe.Uncompress/Transmogrify: Add parent class and Uncompress pipe.

Friday 14 June 2019

2019-06-14 17:44:05 (3 days ago) by Stephen R. van den Berg <srb@cuci.nl>

Shuffler: Memory footprint for fd-shuffling cut in half.

2019-06-14 16:16:33 (3 days ago) by Stephen R. van den Berg <srb@cuci.nl>

Gz: Put classes in separate files to lower memory footprint.

2019-06-14 16:10:19 (3 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Stdio.Buffer: Stricter types.

Use string(8bit) in more places.

2019-06-14 14:07:25 (3 days ago) by Stephen R. van den Berg <srb@cuci.nl>

Gz.Pipe.Compress: New non-blocking compression pipe.

2019-06-14 13:50:17 (3 days ago) by Stephen R. van den Berg <srb@cuci.nl>

Shuffler.Shuffle: Numerous fixes and add set_wrap_callback().

The fixes include:
- Fixed broken end-of-stream detection for d_source_pikestream.c.
- Reduce memory footprint of source-structs by half.
- Fixed memory corruption issues because a memcpy was used on
overlapping regions.
- Eliminated all dynamic memory allocations in the shuffler.
For file-descriptor based streams it still has a fixed overhead of 16KB
per stream, for all other streams the overhead now is close to zero.

Still TODO:
- Reduce the memory overhead for fd-based streams by half; it uses
double buffering now, it could trivially be reduced to use a single
buffer.
HTTP.Server.Request: Use Shuffler to send and support sending chunked data.
Web.Api.Google: Catch errors, then pass them on to the callback.

This is a bit of a kludge, the API should be streamlined with regard
to where and when exceptions are thrown. This currently is done
in a slightly haphazard way.

2019-06-14 13:44:54 (3 days ago) by Stephen R. van den Berg <srb@cuci.nl>

FakePipe: Simplify callbacks.
WebSocket: Reduce systemcall count and speed up response.

Thursday 13 June 2019

2019-06-13 17:44:46 (4 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Testsuite [cpp]: Updated __DIR__ test.

2019-06-13 12:00:03 (4 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Cpp: __DIR__ now represents cwd with ".".

Previously it would sometimes be represented with "",
which was a bad idea considering use like

string file_path = __DIR__ "/file";

2019-06-13 01:14:34 (4 days ago) by Automatic Build <autobuild@roxen.com>

release number bumped to 8.0.828 by export.pike
release number bumped to 8.0.829 by export.pike

Wednesday 12 June 2019

2019-06-12 20:18:16 (5 days ago) by Stephen R. van den Berg <srb@cuci.nl>

FakePipe: Missing close callback invocations added.

2019-06-12 11:31:29 (5 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

release number bumped to 8.0.827 by export.pike
release number bumped to 8.0.826 by export.pike

2019-06-12 11:17:36 (5 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Pike: Move DestructImmediate to the Pike module.

Tuesday 11 June 2019

2019-06-11 20:09:09 (6 days ago) by Stephen R. van den Berg <srb@cuci.nl>

FakePipe: Optimise and allow internal sockets to be destructed independently.

2019-06-11 17:14:06 (6 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Stdio.Buffer: Mark the class for immediate destruction.

This is needed since the object may hold a lock on
a different Stdio.Buffer.

Potentially fixes some of [LysLysKOM 23391165].

2019-06-11 14:22:33 (6 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Compiler: Survive new_program being NULL.

Fixes [LysLysKOM 23386221] some more.

2019-06-11 13:20:20 (6 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Testsuite [Stdio.FakePipe]: Updated to new API.

2019-06-11 12:46:59 (6 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Compiler: Increased paranoia regarding inherit_specifiers.

The value in the variable 'state' is likely to be invalid when
the loop has failed to find the inherit, so do not propagate it.

Thanks to Stephen R. van den Berg <srb@cuci.nl> for the report.

Potential fix for [LysLysKOM 23386221].

2019-06-11 00:25:31 (6 days ago) by Stephen R. van den Berg <srb@cuci.nl>

FakePipe: Numerous bugfixes; it actually works now.

Fixes include:
- Callback handling rationalised, fixed and optimised.
- Eliminate double inheritance which caused subtle conflicts.
- Changes the API to use ->other instead of ->get_other().
- Use single references to mux and cond variables.
FakePipe: Make the direction configurable upon creation.

Monday 10 June 2019

2019-06-10 15:50:50 (7 days ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Constants: Added some constants for RSA PSS signatures.

Sunday 09 June 2019

2019-06-09 13:32:10 (1 week ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Standards.X509: Fixed compilation error when X509_VALIDATION_DEBUG.

Saturday 08 June 2019

2019-06-08 13:53:30 (1 week ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Context: get_signature_algorithms() now knows about ed25519.

Friday 07 June 2019

2019-06-07 13:05:27 (1 week ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL: Improved support for Ed25519 certificates.

Thursday 06 June 2019

2019-06-06 12:08:02 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL: Added some support for Ed25519 certificates.

Wednesday 05 June 2019

2019-06-05 14:33:08 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Testsuite [SSL]: Test ECDHE with all supported curves.

Tuesday 04 June 2019

2019-06-04 16:11:55 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Standards.X4509: Added some support for validation of Ed25519 certificates.

2019-06-04 13:26:40 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Crypto.ECC.Curve25519: Added pkcs_eddsa_id().

This is useful for finding the correct curve to use, given
an eddsa_id, as it differs from the corresponding curve_id.

Also updates some RFC references to reference RFC 8410.

Monday 03 June 2019

2019-06-03 17:39:52 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Documentation [Nettle]: Fixed typo in doc.

2019-06-03 01:13:12 (2 weeks ago) by Automatic Build <autobuild@roxen.com>

release number bumped to 8.0.824 by export.pike
release number bumped to 8.0.825 by export.pike

Sunday 02 June 2019

2019-06-02 22:45:40 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL: Enable ECDHE with Curve 25519 by default.

Verified to work with www.google.com after the recent Curve25519 fix.

2019-06-02 16:53:24 (2 weeks ago) by Stephen R. van den Berg <srb@cuci.nl>

Stdio.File: Support openat with a single argument.

Also fixes:
- Avoid a SEGV when supplied with a single argument.
- Synchronise low level implementation with pike level by setting the
default mask to 0777 everywhere.

2019-06-02 16:48:39 (2 weeks ago) by Stephen R. van den Berg <srb@cuci.nl>

Stdio.File: Support openat with a single argument.

Also fixes:
- Avoid a SEGV when supplied with a single argument.
- Synchronise low level implementation with pike level by setting the
default mask to 0777 everywhere.

Saturday 01 June 2019

2019-06-01 16:04:35 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Nettle.Curve25519: Fixed encoding of Point.

The default implemetation of encode() prefixes the value with '\4'
(ANSI x9.62 encoding #4), but this should NOT be done for Curve25519
as there is only the single encoding.

This fixes key-exchange incompatibilities with SSL.

Friday 31 May 2019

2019-05-31 12:40:08 (2 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Constants: Added some support for Ed25519 and Ed448 certificates.

Thursday 30 May 2019

2019-05-30 13:39:24 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Build: Fixed compilation warning.

Wednesday 29 May 2019

2019-05-29 17:11:52 (3 weeks ago) by Stephen R. van den Berg <srb@cuci.nl>

pgsql: Eliminate race for db connections still open at program exit.

2019-05-29 16:59:52 (3 weeks ago) by Stephen R. van den Berg <srb@cuci.nl>

Web.Api.Google: Properly fail promise on failed callbacks.
Web.Api: Add missing protected classifiers.
Web.Api: Add rest args to the callbacks.
pgsql: Eliminate race for db connections still open at program exit.
threads: Fix backtrace typo.

2019-05-29 16:52:00 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Documentation [SSL.Constants]: Updated AutoDoc.

Tuesday 28 May 2019

2019-05-28 16:40:26 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Context: Fixed another place.

2019-05-28 15:38:55 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Session: Missed one place.

2019-05-28 13:38:50 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL: Use SignatureScheme instead of array({Hash,Signature}Algorithm).

2019-05-28 12:32:10 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Context: Updated cipher_suite_sort_key() to new HASH_*.

2019-05-28 11:25:54 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL: Renumbered HASH_* in preparation for using SignatureScheme.

Monday 27 May 2019

2019-05-27 17:14:17 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Testsuite [Protocols]: Added some comments.

Sunday 26 May 2019

2019-05-26 19:15:27 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.Constants: Fixed typo.

2019-05-26 12:48:40 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.constants: Added some entries from RFC 8422.

Saturday 25 May 2019

2019-05-25 16:10:40 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Standards.PKCS: Updated with references to RFC 8410.

Friday 24 May 2019

2019-05-24 16:42:02 (3 weeks ago) by Stephen R. van den Berg <srb@cuci.nl>

Web.Api: Add and rewrite support for all Google APIs (past/present/future).

2019-05-24 16:26:52 (3 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL: Hide experimental cipher suites.

Thursday 23 May 2019

2019-05-23 16:36:08 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Concurrent.Future: Added query_{success,failure}_callbacks().

2019-05-23 01:13:47 (4 weeks ago) by Automatic Build <autobuild@roxen.com>

release number bumped to 8.0.823 by export.pike
release number bumped to 8.0.822 by export.pike

Wednesday 22 May 2019

2019-05-22 17:42:29 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Tools.Standalone.precompile: Improved doc.

2019-05-22 14:29:52 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Image.Image: Fixed stack underflow in create(..., "grey").

2019-05-22 11:26:18 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

release number bumped to 8.0.820 by export.pike
release number bumped to 8.0.821 by export.pike

2019-05-22 10:53:32 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch 'patches/pike189' into 8.0

* patches/pike189:
Thread.MutexKey: Added _sprintf().

2019-05-22 00:44:56 (4 weeks ago) by Tobias S. Josefowitz <tobij@tobij.de>

Merge branch 'tobij/fix-hilfe-output'

* tobij/fix-hilfe-output:
Hilfe: send_output(): Don't process fmt when called with one arg

2019-05-22 00:28:18 (4 weeks ago) by Tobias S. Josefowitz <tobij@tobij.de>

Hilfe: send_output(): Don't process fmt when called with one arg

In GenericAsyncHilfe, send_output() should not process the input string
as format string when called with one argument only, as that input may
contain percent signs that should not be interpreted (and will cause
sprintf() to throw because it's called with no additional arguments
and/or nonsensical formats).

Tuesday 21 May 2019

2019-05-21 17:22:12 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Ignore the generated pike_types.c file.

2019-05-21 17:18:33 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Types: Use the precompiler to build the type checker.

Converts pike_types.c into pike_types.cmod.

2019-05-21 14:03:31 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Tools.Testsuite: Improved output from log_msg().

2019-05-21 01:13:38 (4 weeks ago) by Automatic Build <autobuild@roxen.com>

release number bumped to 8.0.819 by export.pike
release number bumped to 8.0.818 by export.pike

Monday 20 May 2019

2019-05-20 15:05:06 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

release number bumped to 8.0.816 by export.pike
release number bumped to 8.0.817 by export.pike

2019-05-20 12:37:31 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch 'patches/pike189'

* patches/pike189:
Thread.MutexKey: Added _sprintf().

2019-05-20 12:33:29 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Thread.MutexKey: Added _sprintf().

Fixes [PIKE-189].

2019-05-20 12:09:06 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Protocols.DNS: Survive Stdio.UDP()->bind() failing with EACCES.

This can apparently happen for usually unprivileged ports on
both NT and Solaris 11.

Fixes [PIKE-185].

Sunday 19 May 2019

2019-05-19 20:39:17 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Crypto.ECC.Curve.ECDSA: Improved behavior of equal().

The objects being compared no longer need to be created
from the exact same parent object.

Saturday 18 May 2019

2019-05-18 17:19:07 (4 weeks ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Calendar.mkexpert: Added some notes.

Friday 17 May 2019

2019-05-17 18:44:18 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

SSL.ServerConnection: Improved diag on cipher suite mismatch.

Thursday 16 May 2019

2019-05-16 14:17:53 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

mktreeopt: Match against more specific rules before generic ones.

Potential fix for [CID 1294685], where a generic rule for F_ADD_EQ
(and F_SUB_EQ) shadowed a special case rule.

Wednesday 15 May 2019

2019-05-15 15:48:40 (1 month ago) by William Welliver <william@welliver.org>

Merge branch 'arne/BSON_fixes'

Better bounds checking and test fixes

2019-05-15 15:33:06 (1 month ago) by William Welliver <william@welliver.org>

BSON: fix document length in test

2019-05-15 12:10:40 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch 'patches/pike185'

* patches/pike185:
Protocols.DNS: Survive Stdio.UDP()->bind() failing with EACCES.

2019-05-15 12:08:33 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Protocols.DNS: Survive Stdio.UDP()->bind() failing with EACCES.

This can apparently happen for usually unprivileged ports on
both NT and Solaris 11.

Fixes [PIKE-185].

Tuesday 14 May 2019

2019-05-14 15:20:59 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Tools.sed: Protected lfuns.

2019-05-14 10:54:50 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

MANIFEST: Added some entries.

Monday 13 May 2019

2019-05-13 11:29:38 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Testsuite: Removed obsolete AC_CONFIG_HEADERS test.

AC_CONFIG_HEADERS is now handled via make_variables.

Sunday 12 May 2019

2019-05-12 23:43:44 (1 month ago) by Tobias S. Josefowitz <tobij@tobij.de>

EFUNs: Add deprecated_typep() efun.
EFUNs: Add typeof_identifier()

typeof_identifier(prog, identifier) returns the type of a named
identifier in the program.
Master: Warn about using deprecated members ...

... in join- and dirnodes

The compiler itself does not know the type of the identifiers and will
thus not be able to issue a deprecation warning on its own, thus the
master steps in.
Compiler: Export state of no_deprecation_warnings

... in the active compilation to Pike.
Master: Allow compile warnings with specified lineno
Compiler: export current_{line,file}

2019-05-12 15:59:27 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Build: Removed redundant code from Makefile.in.

The assignment of CONFIG_HEADERS is now done by make_variables.

Saturday 11 May 2019

2019-05-11 15:23:43 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Build: Depend all module objects on the config header (if any).

Fixes cases where the automatic generation of dependencies fails.

Friday 10 May 2019

2019-05-10 15:35:18 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

run_autoconfig: Generate stamp-h.in also when no AC_CONFIG_HEADER.

stamp-h.in is also used as a stamp on the generation of
the configure script.

2019-05-10 13:16:26 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Build: Added some rules for creation of stamp-h.in.

Thursday 09 May 2019

2019-05-09 16:38:06 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Build [MsgPack]: Ignore stamp-h{,.in}.

2019-05-09 16:33:58 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Build: Create the template files for CONFIG_HEADERS on demand.

Wednesday 08 May 2019

2019-05-08 17:10:36 (1 month ago) by Henrik Grubbström (Grubba) <grubba@grubba.org>

Tools.Markdown: Protected lfuns.

[permalink]

Bugs mentioned

  817RESOLVEDSomewhat more complete translation of the admin menu
  821RESOLVED<vinput> doesn't work with the <input> arguments size= and maxlength=
  825RESOLVED<vinput size="10"> does not work
  826RESOLVED<vform enctype="multipart/form-data"> does not work
  827RESOLVEDOracle: <sqltable/> gives backtrace if result set contains null values
  828RESOLVED<emit source="sql"> behaves weird or crashes Roxen with Oracle DB
  829RESOLVEDFiles uploaded with multipart/form-data are split at '\0' by Java Servlet bridge
Bugs? Suggestions?