path: root/sc
AgeCommit message (Collapse)AuthorFilesLines
2013-08-19Resolves: fdo#68220 wrong index into source tableCaolán McNamara1-1/+1
code in master is all very different, so 4-1-* only I think Change-Id: I9277faa462b66a38c7da3ba9a92b0381c67cfe23 Reviewed-on: Reviewed-by: Kohei Yoshida <> Tested-by: Kohei Yoshida <> (cherry picked from commit 48cfdaa4fa453af03d18832e3765ffd65a5e107f) Reviewed-on: Reviewed-by: Eike Rathke <> Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-08-16resolved fdo#67571 write flags and formula token array againEike Rathke1-33/+33
Regression introduced with 337066eb1fa395990b78d1032c399a4688d2a6f9, flags and formula token array weren't written if the formula result was an error value. Change-Id: I8ccd55a7b43d5ceb79002e01f0b3d33d36d85538 (cherry picked from commit 25763e59625ce83de4b82927359108f9e7878744) Reviewed-on: Reviewed-by: Fridrich Strba <> Reviewed-by: Petr Mladek <> Tested-by: Petr Mladek <> Reviewed-by: Kohei Yoshida <> Tested-by: Kohei Yoshida <>
2013-08-06fix build with debug STLDavid Tardon1-7/+17
Failed with: /usr/include/c++/4.8.1/debug/safe_iterator.h:510:error: attempt to compare iterators from different sequences. Objects involved in the operation: ... The problem is that miPos in _copied_ object points ot maAttrs in the original object, not in the copy (and std::for_each takes a copy of the functor). This could be solved by defining copy constructor and operator=, but given the limited usage of the class, it is simpler to let copies share the state. (cherry picked from commit 95d1b05430d0d6384c6910fa9dfdd3c703201d34) Conflicts: sc/source/core/data/documentimport.cxx Change-Id: Icf3f02ecd2fe4ce6dd77f3cde226d32beb4d4b3f Signed-off-by: Stephan Bergmann <>
2013-08-05fdo#67684: Store imported text style entries, else they won't get exported.Kohei Yoshida1-0/+5
Change-Id: Id784772e677e176622e75ff5ae33ab3eb78b1d9a Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-08-05fdo#67099: Don't use edit engine for a single unformatted paragraph.Kohei Yoshida2-8/+38
which is most common, and using edit engine for it would decrease loading performance. Change-Id: I65c20eef96c88edd8eb07c73c27716c4f03c4cda Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-08-05fdo#67099: Remove overheads on inserting cells during ods import.Kohei Yoshida10-260/+485
We will switch to using ScDocumentImport to populate the document from import filters, instead of using ScDocument directly. Conflicts: sc/inc/column.hxx sc/inc/documentimport.hxx sc/source/core/data/column2.cxx sc/source/core/data/column3.cxx sc/source/core/data/documentimport.cxx sc/source/filter/xml/xmlcelli.cxx and a whole bunch of changes backported in order retrofit this code in the 4.1 branch. Change-Id: Ie59d6877d1ac4fc04751a84b663772a9dc9a3efc Reviewed-on: Reviewed-by: Michael Meeks <> Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-08-031st step fix for exporting SheetProtection fdo#64753 is to import correctlyNoel Power6-6/+117
(cherry picked from commit 016e18680e471d9f0cea8ee35f9716ae0f9baa11) 2nd step to export SheetProtection fdo#64753, actually export the data (cherry picked from commit 8ce1124359783df750d176b8390b4eb0f54fb6ba) unit test for fdo#64753 ( import/export of sheetprotection ) Conflicts: sc/qa/unit/subsequent_export-test.cxx Change-Id: If420b386543c15c7b550a720b6e2db69bf372d7f Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-08-02Fix build.Thorsten Behrens1-1/+1
Change-Id: Ic1a6ce26e84982a617d9814645a7bfc3ccd3a63f
2013-08-02fdo#67489: Don't stop range search even when the first cell is empty.Kohei Yoshida2-2/+37
Because we aren't supposed to. Think of those formula cells that follow the empty one. Change-Id: I2cd04833d5e6c561e9ab6d83a984dbdceefa2e71 Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-08-02fdo#67621: Delay populating the ranges until the dialog is shown.Kohei Yoshida2-5/+15
Otherwise the table widget won't have any idea which rows are displayed, and end up not populating the ranges. This is caused by the switch to the .ui dialog, which delays calculation of the dialog size until late. But the code that populates the ranges was execuated when the widget was instantiated, at which time the widget didn't have its size assigned. Change-Id: I22943b41b21b58cc67e872dff13ad3c25eee8438 (cherry picked from commit b8b807ae589a0b0daa101a0b00bcbe21aa138e2f) Reviewed-on: Reviewed-by: Caolán McNamara <> Tested-by: Caolán McNamara <>
2013-07-28fdo#61201: Don't forget to calculate row offset for external single ref.Kohei Yoshida1-0/+1
Change-Id: Ie3dbf2eeba98a60aa2426b1db35ac885102d3821 (cherry picked from commit 8b96cfd6caedbad7b3b79e57421a834f18c5c511) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-28fdo#67246: Detect BIFF 2 (and 3) file format like we should.Kohei Yoshida1-4/+14
(cherry picked from commit b46688a663b8709e0e0795f25ef8961db1f46cba) Change-Id: I1421cca4b0ef8e9410aab5725cc5a8d9cffef7a9 Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-26fix for fdo#62729 reference pos can point to non existing tableNoel Power3-0/+31
there is an existing function ( called at least from uno names api also ) that updates the tab pos, calling that seems to fix this problem (cherry picked from commit b5fffdb8d0438a2fe933a5742d41fe50a14b71f3) with squash of unit test for fdo#62729 (cherry picked from commit 7b3d8e0a7dcf6ae05e1de5c33ed382822cf52cce) Change-Id: I6f6f31895eda9c338eeabd3f3285bf2c9eb23b7e b9be75459aa49b8bab968dedae9e0760ccef9a26 Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-07-26targetted clean of redundant header piece from 62badf3828Michael Meeks20-60/+0
Change-Id: Ic1240114d667fb7797afae4847427cc889f3cb48 (cherry picked from commit 03993b47c522dcc71f7cd73ee21df3de9650080d)
2013-07-25Propagate change on COLUMN() and ROW() functions on position change.Kohei Yoshida5-0/+64
1) Insert =COLUMN() or =ROW() in arbitrary cell. 2) Have another cell reference that cell. 3) Insert or delete column or row to change the cell that contains COLUMN or ROW function. This will change the result of that cell. 4) Check the result of the 2nd cell that references the first. The value change is not propagated. This commit fixes that. NB: master has a similar but different fix. Due to the difference in cell storage between master and 4.1, I had to devise a different fix for the 4.1 branch. Change-Id: Ib1b730e7a4a70a11b967b88730a68362e061a8a0 Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-25resolved fdo#67249 use ScFieldEditEngine to resolve field contentEike Rathke30-90/+167
ScEditUtil::GetString() iterated over the paragraphs of an EditTextObject where GetText() does not resolve field content but returns the embedded field markers. To resolve field content an ScFieldEditEngine is needed. This makes it necessary to pass an ScDocument* to obtain the ScFieldEditEngine from, or for cases where there is no ScDocument in the context use a static ScFieldEditEngine which unfortunately is not capable of resolving document specific fields of course, such as DOCINFO_TITLE and TABLE. Also added unit test. (cherry picked from commit 1ecdc7aaf661e97a33cf521f553481d79cd26de2) Conflicts: sc/qa/unit/subsequent_filters-test.cxx sc/qa/unit/ucalc.cxx sc/source/core/data/column.cxx sc/source/core/data/column2.cxx sc/source/core/data/column3.cxx sc/source/core/data/dociter.cxx sc/source/core/data/table3.cxx sc/source/core/data/table4.cxx Change-Id: Ife3c23b2fec2514b32303239d276c49869786eb5 Reviewed-on: Reviewed-by: Kohei Yoshida <> Tested-by: Kohei Yoshida <>
2013-07-25unit test for <text:s> in <text:p> and <text:span>, fdo#67094Eike Rathke2-8/+49
Change-Id: I033668dcdcdcc1a5710f2ddacadf9a1a5344638e (cherry picked from commit 038d162b175b62d67a94d4418b3a15a1a382419d) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-24fdo#67206: Don't forget to swap the broadcaster storage when swapping columns.Kohei Yoshida2-8/+45
And a new unit test case to catch this in the future. Change-Id: Ife7d0579849a769c23ace8bfd86a39892253a121 Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-07-24followup fix for bnc#823935 exported xlsx causes errors for mso2007Noel Power1-6/+3
previously we added 'file://' scheme to hyperlink urls detected as SMB but.. this is just how linux detects those windows share urls, on windows those urls are detected simply as FILE urls. So... we need to add the scheme if either FILE or SMB ( basically just relax the patch a little ) Change-Id: Ie9a2b0c9e297ee08352db88ad9a21583978dbe3d (cherry picked from commit 40dd3d6e4ed8ed6a2f1c3f953955f9932c0c95ce) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-23resolved fdo#67094 handle <text:s> in <text:p> and <text:span>Eike Rathke4-2/+138
821521f757569c96ded6004bb2cb0d003481b55b introduced XML_SPAN but removed handling of XML_S repeated U+0020, SPACE Change-Id: Ic1b00c9dbc33c750b9a8cae910b4ca0bed42ab5a (cherry picked from commit be10607d358f7587f10e76084893ceed3a4c9215) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-15sc: fix Date/DateTime ambiguityMichael Stahl1-8/+8
(cherry picked from commit 585d4c165f25a6acb19b5f0d05cbb1623283717f) Change-Id: I99908a73d38b0d4b9919ac9c627b849b1d7de0d8 Signed-off-by: Eike Rathke <> Signed-off-by: Norbert Thiebaud <>
2013-07-15i#108348 API CHANGE: add IsUTC to css.util.DateTime etc.Michael Stahl1-1/+1
Add IsUTC member to: Add new stucts with explicit time zones: Adapt the sax::Converter to read/write timezones, and fix the unit test. Everything else just uses default (no time zone), this commit is just to fix the API. STRUCT: /UCR/com/sun/star/util/DateTime nFields1 = 7 != nFields2 = 8 Registry2 contains 1 more fields STRUCT: /UCR/com/sun/star/util/DateTimeRange nFields1 = 14 != nFields2 = 15 Registry2 contains 1 more fields STRUCT: /UCR/com/sun/star/util/Time nFields1 = 4 != nFields2 = 5 Registry2 contains 1 more fields Conflicts: sc/source/filter/oox/unitconverter.cxx Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <> (cherry picked from commit 652ccbdf3111766fadc379a8cf4650b744e1e19c) i#108348: fix TimeZone -> Timezone in struct names (cherry picked from commit 604aae1fd240254fe851d93dc35b5408bd13296c) Signed-off-by: Michael Stahl <> Signed-off-by: Lionel Elie Mamane <> Signed-off-by: Michael Meeks <> Conflicts: forms/source/xforms/convert.cxx offapi/type_reference/offapi.rdb reportdesign/source/core/sdr/RptObject.cxx sc/source/filter/oox/unitconverter.cxx
2013-07-15fdo#63998: Filtering by page fields to work again.Kohei Yoshida3-116/+201
Change-Id: I267ccb30cfa6347313fa1de96e18731f0e22d57d (cherry picked from commit 76488b00ed75bf7915d6d57af9875c6434e740d6) Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-07-13fdo#64448: Don't get type name from incorrect filter.Kohei Yoshida1-3/+2
This causes wrong format type to get "detected" when testing for a completely different file format type. Change-Id: I230759e2be432beeae2db24c12369ccb4585f334 Reviewed-on: Reviewed-by: Lionel Elie Mamane <> Tested-by: Lionel Elie Mamane <>
2013-07-12fdo#66655: Get GETPIVOTDATA to work again.Kohei Yoshida11-116/+114
1) Compare data field name as it is displayed in the table output. 2) In the result tree, store field member names as strings as displayed in the table output, instead of as ScDPItemData. GETPIVOTDATA operates on displayed cell values and do textural comparison. There is no use storing ScDPItemData which screws up value lookup in the result tree. Change-Id: I31bc03a6800f4fadf2ba1180d1958354d43e8a07 Reviewed-on: Reviewed-by: Michael Meeks <> Tested-by: Michael Meeks <>
2013-07-11WaE: pOldRanges may be used uninitializedIvan Timofeev1-1/+1
Change-Id: I22ff60e30b8c5697f648fcd6b7b22a302c6a2135 (cherry picked from commit 7c7ca0d0028cacad15a4fdc3dc02bc5f038c9f3c) Signed-off-by: Eike Rathke <>
2013-07-10fix for fdo#51990 process external refs for macrosNoel Power1-2/+18
(cherry picked from commit f3a0dd8fbc3ead1fe1ec1faec3377cedaacadd93) and also cherry pick from commit adab0134662562298ff13e7fe564623af4bc33da ) use GetMatrix instead of manually disambiguating External and normal refs and cherry-pick further tweak from commit 35f9cd513d6dd796a5bc54c51df3961a674addcb Change-Id: I4fe1d7240f277b532e11d2fbb5aa18f54f2c4695 4fe1d7240f277b532e11d2fbb5aa18f54f2c4695 Reviewed-on: Reviewed-by: Eike Rathke <> Tested-by: Eike Rathke <>
2013-07-10fix reading fonts from .odp using style::font-name (fdo#65338)Luboš Luňák1-1/+1
b40bcde076f9fabf24810d2520e878d604d99637 made writing .odp use style:font-name and office:font-face-decls, instead of using fo:font-family . But the reading back was broken, as xFontDecls is not set in XMLTextImportPropertyMapper::handleSpecialItem(), so the font data was ignored. And xFontDecls was not set because it's set while reading office:font-face-decls, which is at the top of the xml document, but even before the xml is parsed, the call to SdXMLImport::setTargetDocument() calls GetShapeImport(), which creates XMLShapeImportHelper instance, which calls XMLTextImportHelper::CreateParaExtPropMapper(), and XMLTextImportPropertyMapper is created with rImport.GetFontDecls() still being NULL at that point. And it actually doesn't seem to make any sense to just pass around all the pointers to XMLFontStylesContext, as eventually it's always just the one from SvXMLImport. So simply dump all that and make the one single place that actually uses it (i.e. XMLTextImportPropertyMapper::handleSpecialItem()) refer directly to SvXMLImport::GetFontDecls(). Change-Id: Ib1b3e4b1bcaf87ca3e4703d1cc1563ad6b3f9ce7 Reviewed-on: Reviewed-by: Michael Meeks <> Reviewed-by: Michael Stahl <> Tested-by: Michael Stahl <>
2013-07-03the calculation of the progressbar leads to very slow savingLaurent Godard1-3/+3
Change-Id: Ibba31126161b77b499561c5a74c057a56c07a4a8 Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <> Reviewed-on: Reviewed-by: Markus Mohrhard <> Tested-by: Markus Mohrhard <>
2013-07-03temporarily disable tests known to failChristian Lohmaier1-2/+2
Change-Id: I95c3c1af64b7d8b499c085c345b26d2df7e31965 Reviewed-on: Reviewed-by: Markus Mohrhard <> Tested-by: Markus Mohrhard <>
2013-07-03simplify AddPrintRanges calls at import timeLaurent Godard2-9/+10
- at import time, use directly core functions - honor IsUndoEnabled in uno calls Change-Id: If1e762df0ca9cfa22fa2c0e75518fe892be602b6 Reviewed-on: Reviewed-by: Markus Mohrhard <> Tested-by: Markus Mohrhard <> Reviewed-on:
2013-07-03xls cells have all hard number format, fdo#66506Markus Mohrhard1-0/+1
Change-Id: Ia2dd206bd1e9a91bef0021a9317f0d6486b2f276 (cherry picked from commit f45c47af72369859d1937cb8b46cba370031509a) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-02don't crash during chart copying, fdo#66508Markus Mohrhard1-3/+3
(cherry picked from commit 92207231c3742873ad3b3a6fed2963d1bf924ad3) Change-Id: I3a4806fcae7b830e71cda0ab7e6d5ac6f180d4d3 Reviewed-on: Reviewed-by: Kohei Yoshida <> Tested-by: Kohei Yoshida <>
2013-07-02don't forget the formula cell style during xls/xlsx export, fdo#66242Markus Mohrhard2-3/+46
This reverts parts of commit 64223bda068cc6d82d3c5e359d465014568398bb. Change-Id: I6d5379aaedbf011b39a4cf076f0a43bc8a6cbc93 (cherry picked from commit 93a5b7ee36818d60963e4fbe21f9e6c43c7c5a80) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-07-02resolved fdo#35411 redefined MAXCODE 512 to FORMULA_MAXTOKENS 8192Eike Rathke2-4/+2
Also renamed MAXJUMPCOUNT to FORMULA_MAXJUMPCOUNT but without changing the value as the runtime array size of ocChose depends on it, should be changed before. Eliminated the duplicated and error causing redefinition of both in sc/inc/compiler.hxx Change-Id: I0e87d1439c9564a4f475fcb2870ab51c3b586942 (cherry picked from commit 9c1ca6dca3b553c302a635357e33591605343b99) Reviewed-on: Reviewed-by: Markus Mohrhard <> Tested-by: Markus Mohrhard <>
2013-07-01hide the style boxes to collapse the date entries fullyMarkus Mohrhard1-0/+2
Change-Id: I58f06a57a05ef007afc203c737acfe29df6ea6a3 (cherry picked from commit def32c7e14ad9743e2b55804442be5d596f6c21c) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-01update all style lists when a new style is addedMarkus Mohrhard2-4/+67
Change-Id: Ib404958e2bc0e3bfe075d1c94f348e907ee004c8 (cherry picked from commit aa028cab4a72660c773a9eac0cd8a19a0d655c01) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-01handle localized default style names, fdo#61339Markus Mohrhard2-4/+7
We need to map localized style names during import and export from and to ODF. The default styles are the only localized style names and are not imported with the name written into the file. Change-Id: Ibdc2f750b7a4b7ce6994b22248e237fe95ac638d (cherry picked from commit 5b9bad7482a98f2d0d37c4b75a13292abe653ea3) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-01extract this functionality from the uno codeMarkus Mohrhard7-151/+190
we need this code in our next step for some mappings in the filters code Change-Id: I059d9d05877846215e1a967123dc253de605417e (cherry picked from commit b3aabf223ec7bc2678fd6dfbbbbb79ab7f079e5b) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-01only clean the cond format if we want to write attribs, fdo#62267Markus Mohrhard1-1/+1
Change-Id: I2f4feecb3180b165f6b9b299ecb3dcdbb65f87e3 (cherry picked from commit d4f631e8ceb7f02a18565cf0470c8170215e69f8) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-01use the local setting aware method for the UI, fdo#64754Markus Mohrhard1-11/+17
Change-Id: Ie3117c26b347e1a8996ef9cbcb8b0d6ee561b5b3 (cherry picked from commit a539cf2c09bdd62d1c4a0124ed536f88d4fd01d5) Reviewed-on: Reviewed-by: Bosdonnat Cedric <> Tested-by: Bosdonnat Cedric <>
2013-07-01Resolves: fdo#66263 calc's protect records doesn't do anything, sid collisionCaolán McNamara1-1/+2
Change-Id: I50c9f0bd28f17e5d56c6d6948871042ba64650ef (cherry picked from commit 257ba103fb9ffae87189eb5d435969961fe4e98c) Reviewed-on: Reviewed-by: Markus Mohrhard <> Tested-by: Markus Mohrhard <>
2013-06-28we don't need this hack without inherited number formats, fdo#66076Markus Mohrhard6-45/+2
Change-Id: If315ce28a4182019002141c7330475e0843f9a40 (cherry picked from commit 3f0b5770790dfcb408d169df1265546d6e8e3ece) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-06-28unit test for commit:1680a8cd4f4393ec15c4f85cb63b6654117d56d1 (fdo#56960)Noel Power3-3/+22
expand the existing borders import check to cater for xlsx ( and testing for fine/hair border styles ) Change-Id: Ic5c3cd46e0049dbce9997f0881c4d98e0d64e00c (cherry picked from commit 69dcacedaccd1d698a54eed615ee8739747bc975) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-06-28fix hair & fine cell border export ( followon fix fdo#56960 )Noel Power5-10/+11
1) align excel binary and oox import filter border thickness values 2) use new fine-dash border style 3) detect & determine hair vrs fine borderline on export also squashed relevant part of commit ee51444ed1f7003dafc93c8181b5f8c1b0fd165b ( fix borders unit test ( test values have changed ) ) Change-Id: Ieb024bc6f4bfc87c525082c7b9a1f3c444f5c5ca (cherry picked from commit 1680a8cd4f4393ec15c4f85cb63b6654117d56d1) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-06-28fix for bnc#823935Noel Power1-0/+6
window network drive urls on import get internall converted to SMB (so it appears anyway) when exported the file:/// that was on the target ( in the relId ) for the hyperlink is dropped. Apparently the scheme isn't necessary for the binary import. If one of these strange 'detected as smb' window shares are exported make sure we add the file:/// (cherry picked from commit 8e33e95500c2f5edc91e853024a5c0fff285fcce) Change-Id: Id62e1f4845bfbc610f73499881dc33d3d955923b (cherry picked from commit 732640c353b39cff7a9d5b5906f8975102cbed11) Reviewed-on: Reviewed-by: Fridrich Strba <> Tested-by: Fridrich Strba <>
2013-06-27fdo#65856 allow shift cells in paste special for copy-modeWinfried Donkers1-1/+1
Change-Id: I7a7e13104e7256da9a82e16e14ddfa04e35e7f8a Reviewed-on: Tested-by: Eike Rathke <> Reviewed-by: Eike Rathke <> (cherry picked from commit e513fffeac1f4a814f644f156d88fc8810af4e46) Signed-off-by: Eike Rathke <>
2013-06-24clang, -Wunused-private-fieldJulien Nabet1-1/+0
Change-Id: I8aa3d67c50a7b248cfdefb9a3152afd259b8ebbd (cherry picked from commit ac1cb18b82bbee3a2a54ce73480fdb3e2cbf4ab4) Reviewed-on: Reviewed-by: David Tardon <> Tested-by: David Tardon <>
2013-06-17resolved fdo#65082 RATE function should not find roots <= -1Eike Rathke1-13/+46
Limit RATE to roots > -1.0, which is what also Excel and now Gnumeric do. If the Newton goal seek fails for the default guess value 0.1 try other values. This now also calculates the few remaining error cases of i#15090 attachment correctly. Change-Id: Ic62cc807626b4715035c3076f58f303b9020db5a (cherry picked from commit 9ee7be4efb494351c4be096ffa04cdbd85cdc3d4)
2013-06-14ditch dangling reference to Rathke1-3/+0
Change-Id: I185a9854ca2ce43f1f880f5ea2bb3b68a1757ca9 (cherry picked from commit f2fd2a66ee827024b31a310d67804cb7cb18d2da)