summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2023-03-21tdf#153161: (Ab)use a call to XTextRange::getString to flush editscib-6.4-16Mike Kaganski
Restore the call (without checking its returned value), removed in commit d194474aabd699806cb3631bc8641dd0548b8026 ("tdf#151100: xText->getString() may be empty for content needing export", 2022-09-22), side effects of which obviously allow some object's changes to get flushed and saved. Change-Id: I62f27cd056c32ad76f79a4862e2f4a0964eaadef Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146106 Tested-by: Mike Kaganski <mike.kaganski@collabora.com> Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149142 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2023-03-21Bump version to 6.4.0.16Thorsten Behrens
Change-Id: Ib19bb43fcae58bf20df1bf7a1f5863f9cd6215aa
2023-03-21tdf#117948 Do not treat hidden rows as zero in ODF exportRegina Henschel
The object geometry in ODF file format has values so as if no hidden columns or rows exists. But for rendering the object geometry has to treat hidden rows and columns as zero. This patch changes the object geometry temporarily to the 'no hidden columns and rows' mode for export and restores the original geometry afterwards. The patch considers hidden columns left from the shape and hidden rows above the shape. So the object is shifted. Considering hidden columns or rows in the area, which is covered by the shape, is still missing. That would possibly require scaling. cherry-pick from commit: 1cb6bb9576871ff5d56c9810ff1791abc3dd64fd (tdf#117948 Do not treat hidden rows as zero in ODF export) Change-Id: Icdb3f08404ca4d212d25a1967bfdc0bfc7186007 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105427 Tested-by: Jenkins Reviewed-by: Regina Henschel <rb.henschel@t-online.de> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149163 Tested-by: Gabor Kelemen <kelemeng@ubuntu.com> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2023-03-16Check iframe target for allowed document URLsSamuel Mehrbrodt
Change-Id: I00e4192becbc160282a43ab89dcd269f3d1012d8 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147919 Tested-by: Jenkins Reviewed-by: Samuel Mehrbrodt <samuel.mehrbrodt@allotropia.de> (cherry picked from commit 288c0920a8475f9f2c537212e04aa7649192ad8c) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148230 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2023-03-16tdf#154005 sc ods fileopen: fix dropdown form control sizeBalazs Varga
Dropdown form control size was increased by the size of hidden rows or columns. Regression from commit: 1f0b3c7a40edfa81bbc7a58d123a6a2dfd83e4ca (Improve 'resize with cell' handling) Change-Id: Ic903a488cab22286f95cfdf4ee559013fd7bfa02 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148738 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148874
2023-03-03crashtesting: assert on loading forum-mso-en-11942.docxCaolán McNamara
with getToken index past end Change-Id: I9bfeb9cacd8ad89de8008d37c88d15350ef84fb3 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136340 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2023-03-03crashtesting: export of fdo50613-3.odt to docx crashesCaolán McNamara
there are multiple objects in the doc and the ole cache hovers at the max size at which it wants to purge excess object. On save to docx, object a's chart model is fetched in order to save it, this triggers activating that object a, so object b is purged from the cache. storeOwn is called on b to save it to the document persist storage. During save of object b to document persist, ChartView::getExplicitValuesForAxis is called, which calls impl_updateView which eventually calls lcl_setDefaultWritingMode. *if* IsCTLFontEnabled is on, then chart digs through its parents objects looking for a chart in order to see what the writing direction is at the insertion point of the chart, or failing that for the page its on) (this also seems dubious as it might be any chart, not the current chart) To see if the object is a chart it calls getPropertyValue("CLSID") which brings the object into the ole cache, another object is purged from the cache, and the object purged is object b (which is in progress of getting purged already) object b is now purged in the inner case, so when control is returned to the outer storeOwn the object properties have been deleted and all is lost disallow a purge within a purge Change-Id: Ia21e794759aa82b6bcf39c638be8b47ac58a9bb3 Reviewed-on: https://gerrit.libreoffice.org/83808 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Caolán McNamara <caolanm@redhat.com>
2023-03-02tdf#149548 sw: don't rely on binary search in SplitRedline()Michael Stahl
The problem is that for this bugdoc overlapping redlines are created by writerfilter, and so this one survives the split in SwDoc::TextToTable() when creating a cell on node 146 to 149: $125 = (SwRangeRedline) { <SwPaM> = SwPaM = { point = SwPosition (node 155, offset 156), mark = SwPosition (node 146, offset 0) }, It's not found in GetRedline(), as that relies on the redlines never overlapping, so isn't split at end of 149. Then when copying to the clipboard, all redlines are deleted and this deletes cell start and end nodes, while the cell's SwTableBox is still referenced elsewhere. (reportedly somehow a regression from commit c4cf85766453982f1aa94a7f2cb22af19ed100be) Change-Id: I6466e6777cd83c7625381f7049f30c8a2f487af1 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137110 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
2023-03-02tdf#145621 sw: fix crash pasting paragraph with cross-reference markMichael Stahl
At last, a case where copying a mark fails for legitimate reasons. (regression from 4bf04dea9afb30a9395e80b07a81d1908937ee8b) Change-Id: I7a318e03c0b0d30e204eadb5012f190edbbba486 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125767 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
2023-03-01curl: upgrade to release 7.88.1Taichi Haradaguchi
Fixes CVE-2023-23916, 2 CVEs that probably don't affect LO. Reference: https://curl.se/docs/security.html Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147977 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 6074d16b8c631e679a67364837d4ca9799731152) Change-Id: If9b3fc7c5ce66bfe1027caff39ea2c1cf55df7ad
2023-03-01curl: upgrade to release 7.87.0Taichi Haradaguchi
Fixes CVE-2022-43551 and CVE-2022-43552. https://curl.se/docs/CVE-2022-43551.html https://curl.se/docs/CVE-2022-43552.html Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145116 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 181806115a694ade32c7bba1abd9aa931b1a93b9) Change-Id: I979ed11c212aef226ad9f26420462e5f9dbe15e5
2023-03-01postgresql: upgrade to release 13.10Taichi Haradaguchi
Fixes CVE-2022-41862 Reference: https://www.postgresql.org/support/security/CVE-2022-41862/ Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147834 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit efff34d7bdccf210a5e878bd9bc4d35ec72b71ab) Change-Id: I6075838972fec1c091f3150b19c5da4dc80ad6d3
2023-03-01Always push a result, even if it's only an errorEike Rathke
PERCENTILE() and QUARTILE() if an error was passed as argument (or an error encountered during obtaining arguments) omitted to push an error result, only setting the error. Fallout from commit f336f63da900d76c2bf6e5690f1c8a7bd15a0aa2 CommitDate: Thu Mar 3 16:28:59 2016 +0000 tdf#94635 Add FORECAST.ETS functions to Calc Change-Id: I23e276fb0ce735cfd6383cc963446499dcf819f4 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147922 Reviewed-by: Eike Rathke <erack@redhat.com> Tested-by: Jenkins (cherry picked from commit 64914560e279c71ff1233f4bab851e2a292797e6)
2023-02-27tdf#114973 sw: don't allow every body SwTextFrame to be hiddenMichael Stahl
Check this in SwTextFrame::IsHiddenNow() for the first one in the body. Also fix a bad entry in SwScriptInfo::m_HiddenChg that is then possible because it wasn't called on empty frames before. Change-Id: If57f8d1a30229cfbf97225b386365ae384c5083c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147347 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit b1814b7cc851c4346f6cfa81c5be74c46f42e90f)
2023-02-27tdf#114973 sw: enable SelectAll with hidden para at start/endMichael Stahl
If there's a hidden para the shell cursor can't be positioned in it ordinarily, so SelectAll will exclude these at the start or end of the document. There is already special code to handle a table at the start of the document body, and it's relatively simple to adapt it to handle hidden paragraphs as well. This appears to work surprisingly well, the point is at the start of the first node of the document, and moving it right immediately puts it to the first non-hidden paragraph. But it only works for paragraphs hidden by character formatting or hidden paragraph field, not if there's a hidden section - there are no (not even 0-height) SwTextFrames in hidden sections. Change-Id: Ifd3c11f4169a037fdae2c2b376d0138bec46774f Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146257 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit bb733957dd39e6f0b9d80bb59eb0177188794797)
2023-02-27tdf#139566: Fix select all (table + text) when document window looses focus.Tamás Zolnai
Without setting mbSelectAll properly the selection cursor is converted to a SwShellTableCursor, which can't handle if a selection is extended outside of the table. Change-Id: Ibd43097ac782b9fc02ff818d3911ebcd20efbd21 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/109185 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/109205 Tested-by: Jenkins (cherry picked from commit 68b1ae7b493eea4d77f36a0c8566ca0224f04798)
2023-02-27tdf#114973 sw: IsSelOvr() should not leave the textMichael Stahl
Jumping from body to header is certainly unexpected, call FindNextCnt() which does additional checks to stay in the same text environment. Change-Id: If54bff62cac420dc0990b6be070b81fa56365c50 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146256 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 5aeee226479427062ac885be407671981d6940ef) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146205 Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit a3cd59aa9e6ea3dfbfb725201ca143c3ec29aa81)
2023-02-26tdf#134426 tdf#138873 sw: Revert "tdf#79717 ...Justin Luth
...save/restore character format on selection overwrite" which might also fix tdf#144638 as well. This reverts LO 6.3 commit 6abed0ea006f3616e40faf2ae782cf64f8ac2914. This 6.3 fix for bug 79717 has caused too many other problems that are not being looked at, so it should just be reverted. It is spamming documents with direct formatting which is a terrible thing to have happen by accident. It completely wrecks the proper use of styles. Also revert asscoiated follow-up commits: Revert "tdf#79717 save/restore character style on selection overwrite" This reverts commit 6.3 04bd1925706360414438b814046b543c5e317d0a. Revert "tdf#79717: sw_uiwriter: Add unittest" This reverts commit 7.1 b05955b480fe4d32852e7be8a118d46ca7e6dbfa. These effectively revert 7.0.4 12eac5bcbc9b71bf00cb88b918988826229cca35 but its unit test is left intact. Change-Id: Ideced4d38bfdf4c82f1744534afbaad29689fded Reviewed-on: https://gerrit.libreoffice.org/c/core/+/123566 Tested-by: Jenkins Reviewed-by: Justin Luth <justin_luth@sil.org> Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/123577 Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147577 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2023-02-21nss: upgrade to release 3.88.1cib-6.4-15Michael Stahl
Fixes CVE-2023-0767 CVE-2022-3479 Change-Id: I688dc7d0785ed3344c33e331c7e9ef37baa720ee Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147387 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 538975a0e511ad79a7dd3c71300b993d1554cd03)
2023-02-21nss: upgrade to release 3.86.0Taichi Haradaguchi
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/143038 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit a2969884af71e88d2541bd2bfdebea222876fef4) Change-Id: Ia236c7124d920785f7a2856db5ee1ccbef7a2d68
2023-02-21nss: update to release 3.83Taichi Haradaguchi
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140231 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 06c9e1d27093ca7bbe38a1eb93fbb85b80f75f75) Change-Id: Ic1cb36b3a56cde40296fd46ef356dea286932692
2023-02-20Stack check safety belt before fishing in muddy watersEike Rathke
Have it hit hard in debug builds. Change-Id: I9ea54844a0661fd7a75616a2876983a74b2d5bad Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147205 Reviewed-by: Eike Rathke <erack@redhat.com> Tested-by: Jenkins (cherry picked from commit 9d91fbba6f374fa1c10b38eae003da89bd4e6d4b) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147245 Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 166a07062dd4ffedca6106f439a6fcddaeee5eb5)
2023-02-20Obtain actual 0-parameter count for OR(), AND() and 1-parameter functionsEike Rathke
OR and AND for legacy infix notation are classified as binary operators but in fact are functions with parameter count. In case no argument is supplied, GetByte() returns 0 and for that case the implicit binary operator 2 parameters were wrongly assumed. Similar for functions expecting 1 parameter, without argument 1 was assumed. For "real" unary and binary operators the compiler already checks parameters. Omit OR and AND and 1-parameter functions from this implicit assumption and return the actual 0 count. Change-Id: Ie05398c112a98021ac2875cf7b6de994aee9d882 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147173 Reviewed-by: Eike Rathke <erack@redhat.com> Tested-by: Jenkins (cherry picked from commit e7ce9bddadb2db222eaa5f594ef1de2e36d57e5c) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147129 Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit d6599a2af131994487d2d9223a4fd32a8c3ddc49)
2023-02-15disable script dumpCaolán McNamara
Change-Id: I04d740cc0fcf87daa192a0a6af34138278043a19 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146986 Tested-by: Jenkins Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147051 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2023-02-12Bump version to 6.4.0.15Thorsten Behrens
Change-Id: I41bbe7003e8caa2905b9434ae4f3a0169cc22df6
2023-02-08openssl: upgrade to release 1.1.1tMichael Stahl
Fixes CVE-2023-0286 CVE-2023-0215 CVE-2022-4450 CVE-2022-4304 Change-Id: I93ce0362b17bd07b0644564a0676daaa56bc8b50 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146653 Tested-by: Jenkins Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de> (cherry picked from commit f9229fdadbd205a8953577efc72a6e43717c764e)
2023-02-08add three static lines beween tarballs to ease cherry-pickingMichael Stahl
Change-Id: I408f7cd971350d1c201560f7fb5da5d2bc78e0d4
2023-02-08don't export entries from download.lst - only used by makeChristian Lohmaier
no need to plant that into the environment of every recipe. some Jenkins Windows builders already run into "environment is too large for exec" issue just by attempting a "xargs --show-limits </dev/null" from a dummy rule within the make environment Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144217 Tested-by: Jenkins Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com> (cherry picked from commit ec4409a911fd4a67eb1bc74aae33081ae8510339) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144325 Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 70f0cf457bdd36e914f3e98cf027ee25a75df448) Cherry-pick part of commit d9617d5529e0da23bd8602d07ce459cc42746e02 to fix make module.clean. Change-Id: I321fa9075532eef62a7d4e33a08c272276de717e
2023-01-09tdf#131046 Add back Number Format item to Writer table context menusGabor Kelemen
Change-Id: I9b7ed2ebe808f05c778b8a993b4df4fdd4f6a151
2023-01-05tdf#140296: Added Edit Fields(context menu) inside of tablesAyhan Yalçınsoy
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/114986 Tested-by: Jenkins Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org> (cherry picked from commit 1cf47d932580a58a7855fe29ac39d17cdd79ae01) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115091 Reviewed-by: Ayhan Yalçınsoy <ayhanyalcinsoy@pisilinux.org> Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com> Change-Id: I22b6a179c1575171315d9eb40cd4895c35d77b6f
2022-12-27Teach symstore more duplicated DLLscib-6.4-14Thorsten Behrens
Amending commit b4dfba947768834ffecc09056992019878711c8b with adding the copied DLLs here, too (symstore complains otherwise) Change-Id: If5cc155cf2fe11b0f97cd152993609fd2c835316 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120686 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2022-12-26Bump version to 6.4.0.14Thorsten Behrens
Change-Id: I4768cc2c2bcdb6fb62274a4447e6d16b94fb289c
2022-12-16tdf#149595 sw: fix showing of fly at start/end of single-para deleteMichael Stahl
... with change tracking enabled. Corner case where there are no extents for the node but the deletion is in a single node (i.e. delete from node start to node end). Make this consistent with what happens without change tracking. (regression from commit 28b77c89dfcafae82cf2a6d85731b643ff9290e5) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144109 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 80f4abfea971cf66259b65fe09366d31ed41809d) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144133 Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com> (cherry picked from commit db29e834d1a2859492a61c57fe1a5131c8ea10eb) Change-Id: Ib63c4108c86c36508e1a378824ab3511dc46ae47
2022-12-16(related: tdf#149595) sw: also re-sort SwSortedObjs on SwPageFrameMichael Stahl
... if SwTextNode content changes; for the unit test here we get: Error: elements in iterator range [first, last) are not partitioned by the predicate __comp and value __val. Also, the SwTextNode may have multiple text frames on multiple pages. Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144108 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit c684301352bcca709bc8d1c99e0f08e7096e9716) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144131 Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com> (cherry picked from commit 2033a23379c0eb9d6de41c7b1473769fda2cc632) Change-Id: I05d4276e4b4b94fe3cd2cdfda3418ad51fa08f5f
2022-12-13sw: fix crash in SwAttrSet::CopyToModify()Miklos Vajna
Regression from commit 938a4d6624a78f3e272b3c4c07f314cb0c6db723 (tdf#128375 sw: fix copying RES_PARATR_LIST_AUTOFMT to different SwDoc, 2019-11-01), the problem is that we may get an autofmt pool item which is set, but its style handle is empty. Assume that this is the same case as having no autofmt at all. Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132301 Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Tested-by: Jenkins (cherry picked from commit ddfafd9b67e6e61c144b29ac73bbafb413ddf03b) Change-Id: I87494fd04687d31201b4ec712cb0fb1ec7362b46 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/143888 Tested-by: Thorsten Behrens <thorsten.behrens@allotropia.de> Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
2022-12-09python3: upgrade to release 3.8.16Michael Stahl
Fixes CVE-2022-37454 and a few other ones that look less relevant including CVE-2022-45061 and CVE-2015-20107. Reviewed-on: https://gerrit.libreoffice.org/c/core/+/143849 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 05f55b3898407828bb24347be56247f58803f7bb) Change-Id: I10fd254f7f0801d47119234bb3436874e98d8c91
2022-12-09Python3: update to 3.8.15Taichi Haradaguchi
* Fixes CVE-2022-40674 * Removed 0001-3.6-bpo-17239-Disable-external-entities-in-SAX-parse.patch.1 as fixed upstream Reviewed-on: https://gerrit.libreoffice.org/c/core/+/141691 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit ea5843b67f1c2006aa1f68f2d00a991e1d463262) Change-Id: I8e71f9a6b013ca4c45bf8774b284be98eee71bab
2022-12-09external/python3: Silence UBSan errors with --with-pydebugStephan Bergmann
...that happen when building ExternalProject_python3 itself after 12142490cd43f8568ab29e0ddfa75b334d6d39d5 "Enable Python Py_DEBUG setting when built with --enable-dbgutil on Linux": For one, silence > Modules/posixmodule.c:14395:9: runtime error: left shift of 34 by 26 places cannot be represented in type 'int' > #0 in all_ins at workdir/UnpackedTarball/python3/./Modules/posixmodule.c:14395:9 where at least my kernel-headers-5.9.9-200.fc33.x86_64 /usr/include/linux/memfd.h has > #define MFD_HUGE_16GB HUGETLB_FLAG_ENCODE_16GB and /usr/include/asm-generic/hugetlb_encode.h has > #define HUGETLB_FLAG_ENCODE_16GB (34 << HUGETLB_FLAG_ENCODE_SHIFT) For another (and as predicted in 29d47d22c43e6adc1850b7db5880028dcd07d1b3 "Fix passing --disable-optimized into external/python3": "in a Linux UBsan build, making ExternalProject_python3 would have started to cause some 'applying zero offset to null pointer' failures, but which would have been easy to fix"), silence > Objects/listobject.c:551:24: runtime error: applying zero offset to null pointer > #0 in list_concat at workdir/UnpackedTarball/python3/Objects/listobject.c:551:24 Change-Id: I0523cd35e393000c8e67629a0522b2db1d8c16f0 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/106984 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com> (cherry picked from commit 1657639d5e405c6e1e988d51a1f267c378c74b53)
2022-12-09python3: upgrade to release 3.8.13Taichi Haradaguchi
Remove some code in external/python3/darwin.patch.0 as it was fixed upstream(https://bugs.python.org/issue45405). Reviewed-on: https://gerrit.libreoffice.org/c/core/+/139493 Tested-by: Jenkins Tested-by: Caolán McNamara <caolanm@redhat.com> Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit c8fd50f9ec02a131fd4c99e80f7252b49015653a) Change-Id: Ie6bfb2456f96a63adbf0dbcb9c902dc56f1151ee
2022-12-09python3: update to 3.8.10Jan-Marek Glogowski
So we don't build 3.8.8rc1 anymore. I didn't look into 3.9. Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117757 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de> (cherry picked from commit c22fc8e1f60bb98a87d22e7ff9bd3290dbb9fe02) Change-Id: Ife7d898c913b9b164168b0ef23a055deea55815f
2022-12-09python3: upgrade to release 3.8.8rc1Michael Stahl
Fixes CVE-2021-3177 plus these less important ones: CVE-2021-23336 CVE-2020-27619 CVE-2020-26116 CVE-2019-20907 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/111208 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit a0c8dc42335764d07c16a017c6b00486ec17ae53) Change-Id: Idbe072a9db1faf8363b4f7795b9fde71c26969f0
2022-12-09python3: update to 3.8.4Jan-Marek Glogowski
With all the prerequisites in place, LO can be updated to the current Python release. Interestingly I found that Cygwin always seems to use LC_COLLATE=C, probably because the default collation rules are missing. Then there are the changes introduced in "PEP 587 -- Python Initialization Configuration", which appearingly have modified the DLL search path behaviour on Windows, so the OpenSLL DLLs aren't found anymore in the program directory. As a workaround, the OpenSLL and libffi DLLs are now (also) installed into the Python lib dir on Windows. Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98437 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de> (cherry picked from commit b4dfba947768834ffecc09056992019878711c8b) Change-Id: Ib82f7b77213da9c525f8c79a13d128d9eec9ca64
2022-12-09libffi: build DLL on WindowsJan-Marek Glogowski
The build setup is rather horrible, with some minimal gcc MSVC wrapper. But the DLL is a prerequisite for the Python 3.8 build, which dropped the internal libffi. It's also possible to build it statically, but then you have to patch the Python 3 _ctypes msbuild properties. This also defaults to explicit --build and --host settings, even without a cross build, because the predicted name would otherwise differ (*-unknown-* instead of *-pc-*). Additionally a "make install" also fails... Change-Id: Ifb7dac840e23efffb9a5e342560aef9e11e0db79 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98436 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de> (cherry picked from commit 883068462fe5bcbb01a8e14736fc06d0c3695c62)
2022-12-09fix some more python 3.8 deprecation warningsNoel Grandin
Change-Id: I8bdbf05f1357aea83a3cdda2f06d63c7d04de8f9 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94561 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk> (cherry picked from commit 0a3d414fb0ac27292320d99f802722a8a9670240)
2022-12-09fix python 3.8 deprecation warningsNoel Grandin
the logo changes were caused by > Support of nested sets and set operations as in Unicode Technical Standard > #18 might be added in the future. This would change the syntax, so to facilitate > this change a FutureWarning will be raised in ambiguous cases for the time being. > That includes sets starting with a literal '[' or containing literal character > sequences '--', '&&', '~~', and '||'. > To avoid a warning escape them with a backslash. Change-Id: I4d48be3df2eaadf03a9d1f5750c0c94b3abbf674 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94191 Tested-by: Noel Grandin <noel.grandin@collabora.co.uk> Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk> (cherry picked from commit b5fcdc3c07efb2c1175503b9c70e6d7336aa1452)
2022-12-09upgrade to openssl-1.1.1sXisco Fauli
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/142184 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 23e6227ab347a00b235fb00892b379ef4e3a0d35) Change-Id: Ic0f1fca7ef73b3a443c24d2bcc7f234be331a05b
2022-12-09upgrade openssl-1.1.1qTaichi Haradaguchi
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/139463 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 96db5e3d610ede2ed82f1ab7673ac6d1c69fd588) Change-Id: I2317e734f074cf7301a6081cf3d2221beeaf5ad1
2022-12-09upgrade to openssl-1.1.1mCaolán McNamara
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128769 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 0f2b8305ed31746cd4dca33aa080adb220109353) Change-Id: I7c3e576501b4587a0f6814cdc4895663ef6d3f79
2022-12-09openssl: upgrade to release 1.1.1lMichael Stahl
Fixes CVE-2021-3712, CVE-2021-3711 (not obvious if any of them affect LO) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121026 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@allotropia.de> (cherry picked from commit 5c391f4346e86bd5d7528fbb42a3af64f98a03d3) Change-Id: I98652348977a5a3c728f1d4fdf7293b76a93b630
2022-12-09upgrade to openssl-1.1.1kCaolán McNamara
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120490 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit c6604b69cf7f84146cae541366c023eaff4c59b6) Change-Id: Ibd930ab4f97e2f74868b73163d7f2f46fb466749
generated by cgit (git 2.34.1) at 2024-07-13 16:43:26 +0000