- Fixed NameError. 337
- Remove tempdirs via finalizers on garbage collection. This cleans them up in all known cases, rather than just when the #close method is called. The #close method can be used to cleanup early. 329
- Fixed README.md typo 318
- Parse sheets in ODS files once to improve performance 320
- Fix some Cell conversion issues 324 and 331
- Improved memory performance 332
- Added
no_hyperlinks
option to improve streamig performance 319 and 333
- Roo::Base::TEMP_PREFIX should be accessed via Roo::TEMP_PREFIX
- The private Roo::Base#make_tempdir is now available at the class level in classes that use tempdirs, added via Roo::Tempdir =======
- Discard hiperlinks lookups to allow streaming parsing without loading whole files
- Fixed opening spreadsheets with charts 315
- Fixed memory issues for Roo::Utils.number_to_letter 308
- Fixed Roo::Excelx::Cell::Number to recognize floating point numbers 306
- Fixed version number in Readme.md 304
- Added initial support for HTML formatting 278
- Handle url with long query params (ex. S3 secure url) 302
- Allow streaming for Roo::CSV 297
- Export Fixnums to Added csv 295
- Removed various Ruby warnings 289
- Fix incorrect example result in Readme.md 293
- Properly parse sci-notation number like 1E-3 #288
- Include all tests in default rake run #283
- Fix zero-padded numbers for Excelx #282
- Moved
ERROR_VALUES
from Excelx::Cell::Number ~> Excelx. #280
- Excelx::Cell::Number will return a String instead of an Integer or Float if the cell has an error like #DIV/0, etc. #273
- Excelx::Cell::Number now handles cell errors. #273
- Added support for returning Integers values to Roo::OpenOffice #258
- A missing Header Raises
Roo::HeaderRowNotFoundError
#247 - Roo::Excelx::Shared class to pass shared data to Roo::Excelx sheets #220
- Proper Type support to Roo::Excelx #240
- Added Roo::HeaderRowNotFoundError #247
- Made spelling/grammar corrections in the README260
- Moved Roo::Excelx::Format module #259
- Updated README with details about
Roo::Excelx#each_with_streaming
method #250
- Fixed Base64 not found issue in Open Office #267
- Fixed Regexp to allow method access to cells with multiple digits #255, #268
Fixed invalid new lines with x000D character#231
Fixed missing URI issue. #245
- Added support for Excel 2007
xlsm
files. #232 - Roo::Excelx returns an enumerator when calling each_row_streaming without a block. #224
- Returns an enumerator when calling
each
without a block. #219
- Removed tabs and windows CRLF. #235, #234
- Fixed Regexp to only check for valid URI's when opening a spreadsheet. #229
- Open streams in Roo:Excelx correctly. #222
- Return an enumerator when calling '#each' without a block #219
- Added Roo::Base#close to delete any temp directories#211
- Offset option for excelx #each_row. #214
- Allow Roo::Excelx to open streams #209
- Use gsub instead of tr for double quote escaping #212, #212-patch
- Fixed Changelog links and release data. #204, #206
- Allow Pathnames to be used when opening files. #207
- Removed the scripts folder. #213
- Added optional support for hidden sheets in Excelx and LibreOffice files #177
- Roo::OpenOffice can be used to open encrypted workbooks. #157
- Added streaming for parsing of large Excelx Sheets. #69
- Added Roo::Base#first_last_row_col_for_sheet a0dd800
- Added Roo::Base#collect_last_row_col_for_sheet a0dd800
- Added Roo::Base::MAX_ROW_COL, Roo::Base::MIN_ROW_COL a0dd800
- Extract Roo::Font to replace equivalent uses in Excelx and OpenOffice. 23e19de
- Roo::Utils 3169a0e
- Roo::ExcelxComments 0a43341 0a43341
- Roo::Excelx::Relationships 0a43341
- Roo::Excelx::SheetDoc 0a43341 c2bb7b8 +- Roo::Excelx::Styles c2bb7b8 +- Roo::Excelx::Workbook c2bb7b8
- Switch from Spreadsheet::Link to Roo::Link ee67321
- Roo::Base#to_csv: Added separator parameter (defaults to ",") #102
- Added development development gems #104
- Reduced size of published gem. #194
- Stream the reading of the dimensions #192
- Return
nil
when a querying a cell that doesn't exist (instead of a NoMethodError) #192, #165 - Roo::OpenOffice#formula? now returns a
Boolean
instead of aString
ornil
#191 - Added a less verbose Roo::Base#inspect. It no longer returns the entire object. #188, #186
- Memoize Roo::Utils.split_coordinate #180
- Roo::Base: use regular expressions for extracting headers #173
- Roo::Base: memoized
first_row
/last_row
first_column
/last_column
and changed the default value of thesheet
argument fromnil
todefault_sheet
a0dd800 - Roo::Base: changed the order of arguments for
to_csv
to (filename = nil, separator = ',', sheet = default_sheet) from (filename=nil,sheet=nil) 1e82a21 - In OpenOffice / LibreOffice, load the content xml lazily. Leave the tmpdir open so that reading may take place after initialize. The OS will be responsible for cleaning it up. adb204b
- Lazily initialize @default_sheet, to avoid reading the sheets earlier than necessary. Use the #default_sheet accessor instead. 704e3dc
- Roo::Base#default_sheet is no longer an attr_reader 704e3dc
- In Excelx, load styles, shared strings and the workbook lazily. Leave the tmpdir open so that reading may take place after initialize. The OS will be responsible for cleaning it up. a973237, 4834e20c, e49a1da
- Change the tmpdir prefix from oo_ to roo_ 102d5fc
- Accept the tmpdir_root option in Roo::Excelx 0e325b6
- Refactored Excelx#comment? 0fb90ec
- Refactored Roo::Base#find, #find_by_row, #find_by_conditions. 1ccedab
- Extended Roo::Spreadsheet.open so that it accepts Tempfiles and other arguments responding to
path
. Note they require an :extension option to be declared, as the tempfile mangles the extension. #84.
- Process sheets from Numbers 3.1 xlsx files in the right order. #196, #181, #114
- Fixed comments for xlsx files exported from Google #197
- Fixed Roo::Excelx#celltype to return :link when appropriate.
- Fixed type coercion of ids. #192
- Clean option only removes spaces and control characters instead of removing all characters outside of the ASCII range. #176
- Fixed parse method with
clean
option #184 - Fixed some memory issues.
- Fixed Roo::Utils.number_to_letter #180
- Fixed merged cells return value. Instead of only the top-left cell returning a value, all merged cells return that value instead of returning nil. #171
- Handle headers with brackets #162
- Roo::Base#sheet method was not returning the sheet specified when using either an index or name #160
- Properly process paths with spaces. #142, #121, #94, 4e7d7d1
- Disambiguate #open call in Excelx#extract_file. #125
- Fixed that #parse-ing with a hash of columns not in the document would fail mysteriously. #129
- Fixed Excelx issue when reading hyperlinks #123
- Fixed invalid test case #124
- Fixed error in test helper file_diff 56e2e61
- Stopped
inspect
from being called recursively. #115 - Fixes for Excelx Datetime cells. #104, #120
- Prevent ArgumentError when using
find
#100 - Export to_csv converts link cells to url #93, #108
- Roo::Excel - Extracted to roo-xls gem. a7edbec
- Roo::Excel2003XML - Extracted to roo-xls gem. a7edbec
- Roo::Google - Extracted to roo-google gem. a7edbec
- Roo::OpenOffice::Font - Refactored into Roo::Font
- Removed Roo::OpenOffice.extract_content a74157a
- Removed OpenOffice.process_zipfile 835368e
- Roo::OpenOffice#comment?
- Roo::ZipFile - Removed the Roo::ZipFile abstraction. Roo now depends on rubyzip 1.0.0+ d466950
- SpreadSheet::Worksheet - Extracted to roo-xls gem. a7edbec
- Spreadsheet - Extracted to roo-xls gem. a7edbec
- Fix that Excelx link-cells would blow up if the value wasn't a string. Due to the way Spreadsheet::Link is implemented the link text must be treated as a string. #92
- Fix that Excelx creation could blow up due to nil rels files. #90
- Support extracting link data from Excel and Excelx spreadsheets, via Excel#read_cell() and Excelx#hyperlink(?). #47
- Support setting the Excel Spreadsheet mode via the :mode option. #88
- Support Spreadsheet.open with a declared :extension that includes a leading '.'. #73
- Enable file type detection for URI's with parameters / anchors. #51
- Fix that CSV#each_row could overwrite the filename when run against a uri. #77
- Fix that #to_matrix wasn't respecting the sheet argument. #87
- Support rubyzip >= 1.0.0. #65
- Fix typo in deprecation notices. #63
- Support :boolean fields for CSV export via #cell_to_csv. #59
- Fix that Excelx would error on files with gaps in the numbering of their internal sheet#.xml files. #58
- Fix that Base#info to preserve the original value of #default_sheet. #44
- Rename Openoffice -> OpenOffice, Libreoffice -> LibreOffice, Csv -> CSV, and redirect the old names to the new constants
- Enable Roo::Excel, Excel2003XML, Excelx, OpenOffice to accept an options hash, and deprecate the old method argument based approach to supplying them options
- Roo's roo_rails_helper, aka the
spreadsheet
html-generating view method is currently deprecated with no replacement. If you find it helpful, tell http://github.com/Empact or extract it yourself.
- Add Roo::Excelx#load_xml so that people can customize to their data, e.g. #23
- Enable passing csv_options to Roo::CSV, which are passed through to the underlying CSV call.
- Enable passing options through from Roo::Spreadsheet to any Roo type.
- Enable passing an :extension option to Roo::Spreadsheet.new, which will override the extension detected on in the path #15
- Switch from google-spreadsheet-ruby to google_drive for Roo::Google access #40
- Make all the classes consistent in that #read_cells is only effective if the sheet has not been read.
- Roo::Google supports login via oauth :access_token. #61
- Roo::Excel now exposes its Spreadsheet workbook via #workbook
- Pull #load_xml down into Roo::Base, and use it in Excel2003XML and OpenOffice.
- #formula? now returns truthy or falsey, rather than true/false.
- Base#longest_sheet was moved to Excel, as it only worked under Excel
- Fix that Roo::CSV#parse(headers: true) would blow up. #37
- Fix that Roo::Spreadsheet.open wasn't tolerant to case differences.
- Fix that Roo::Excel2003XML loading was broken #27
- Enable loading Roo::Csv files from uris, just as other file types #31
- Fix that Excelx "m/d/yy h:mm" was improperly being interpreted as date rather than datetime #29
- Exclude test/log/roo.log test log file from the gemspec in order to avoid a rubygems warning: #26
- Support ruby 2.0.0 by replacing Iconv with String#encode #19
- Excelx: Loosen the format detection rules such that more are successfully detected #20
- Delete the roo binary, which was useless and not declared in the gemspec
- Drop support for ruby 1.8.x or lower. Required in order to easily support 2.0.0.
- Support both nokogiri 1.5.5 and 1.5.6 (Karsten Richter) #18
- Relax our nokogiri dependency back to 1.4.0, as we have no particular reason to require a newer version.
- Support opening URIs with query strings https://github.com/Empact/roo/commit/abf94bdb59cabc16d4f7764025e88e3661983525
- Support both http: & https: urls https://github.com/Empact/roo/commit/fc5c5899d96dd5f9fbb68125d0efc8ce9be2c7e1
- forgot dependency 'rubyzip'
- at least one external application does create xlsx-files with different internal file names which differ from the original file names of Excel. Solution: ignore lower-/upper case in file names.
- New class Csv.
- Openoffice, Libreoffice: new method 'labels'
- Excelx: implemented all methods concerning labels
- Openoffice, Excelx: new methods concerning comments (comment, comment? and comments)
- XLSX: some cells were not recognized correctly from a spreadsheet file from a windows mobile phone.
- labels: Moved to a separate methode. There were problems if there was an access to a label before read_cells were called.
- Openoffice: Better way for extracting formula strings, some characters were deleted at the formula string.
- new class Libreoffice (Libreoffice should do exactly the same as the Openoffice class. It's just another name. Technically, Libreoffice is inherited from the Openoffice class with no new methods.
- Openoffice: file type check, deletion of temporary files not in ensure clause
- Cell type :datetime was not handled in the to_csv method
- Better deletion of temporary directories if something went wrong
- Method #formulas moved to generic-spreadsheet class (the Excel version is overwritten because the spreadsheet gem currently does not support formulas.
- Openoffice/Excelx/Google: #formulas of an empty sheet should not result in an error message. Instead it should return an empty array.
- Openoffice/Excelx/Google: #to_yaml of an empty sheet should not result in an error message. Instead it should return an empty string.
- Openoffice/Excelx/Google: #to_matrix of an empty sheet should not result in an error message. Instead it should return an empty matrix.
- removed gem 'builder'. Functionality goes to gem 'nokogiri'.
- Excel: remove temporary files if spreadsheed-file is not an excel file and an exception was raised
- Excelx: a referenced cell with a string had the content 0.0 not the correct string
- Fixed a problem with a date cell which was not recognized as a Date object (see 2011-05-21 in excelx.rb)
- new method 'to_matrix'
- missing dependencies defined
- double quoting of '"' fixed
- syntax in nokogiri methods
- missing dependency ...rubyzip
- Ruby 1.9 compatible
- oo.aa42 as a shortcut of oo.cell('aa',42)
- oo.aa42('sheet1') as a shortcut of oo.cell('aa',42,'sheet1')
- oo.anton as a reference to a cell labelled 'anton' (or any other label name) (currently only for Openoffice spreadsheets)
- fixed encoding in #cell at exported Google-spreadsheets (.xls)
- added celltype :datetime in Excelx
- added celltype :datetime in Google
- added celltype :datetime in Openoffice and Excel
- Excelx: improved the detection of cell type and conversion into roo types
- All: to_csv: changed boundaries from first_row,1..last_row,last_column to 1,1..last_row,last_column
- All: Environment variable "ROO_TMP" indicate where temporary directories will be created (if not set the default is the current working directory)
- Excel: improved the detection of last_row/last_column (parseexcel-gem bug?)
- Excel/Excelx/Openoffice: temporary directories were not removed at opening a file of the wrong type
- Excel: speed improvements
- Changed the behavior of reading files with the wrong type
- Google: added normalize in set_value method
- Excel: last_row in Excel class did not work properly under some circumstances
- all: fixed a bug in #to_xml if there is an empty sheet
- Excelx: fixed a bug when there are .xml.rels files in the XLSX archive
- Excelx: fixed a bug with celltype recognition (see comment with "2008-07-03")
- Excel: row/column method Fixnum/Float confusion
- support of Excel's new .xlsx file format
- method #to_xml for exporting a spreadsheet to an xml representation
- fixed a bug with excel-spreadsheet character conversion under Macintosh Darwin
- fixed a bug with excel-spreadsheet character conversion under Solaris
- no more tmp directories if an invalid spreadsheet file was openend
- new celltype :time
- time values like '23:15' are handled as seconds from midnight
- additional 'sheet' parameter in Google#set_value
- fixed a bug within Google#set_value. thanks to davecahill [email protected] for the patch.
- better support of roo spreadsheets in rails views
- fixed a bug within #to_cvs and explicit call of a sheet
- fixed 'find_by_condition' for excel sheets (header_line= --> GenericSpredsheet)
- another fix for the encoding issue in excel sheet-names
- reactived the Excel#find method which has been disappeared in the last restructoring, moved to GenericSpreadsheet
- basename() only in method #info
- changed logging-method to mysql-database in test code with AR, table column 'class' => 'class_name'
- reactived the Excel#to_csv method which has been disappeared in the last restructoring
- fixed a bug with first/last-row/column in empty sheet
- #info prints now '- empty -' if a sheet within a document is empty
- tried to fix the iconv conversion problem
- Google online spreadsheets were implemented
- some methods common to more than one class were factored out to the GenericSpreadsheet (virtual) class
- Openoffice/Excel: the most methods can be called with an option 'sheet' parameter which will be used instead of the default sheet
- Excel: improved the speed of CVS output
- Openoffice/Excel: new method #column
- Openoffice/Excel: new method #find
- Openoffice/Excel: new method #info
- better exception if a spreadsheet file does not exist
- Openoffice: percentage-values are now treated as numbers (not strings)
- Openoffice: refactoring
- Openoffice: repeating date-values in a line are now handled correctly
- csv-output to stdout or file
- Openoffice: fixed a bug with internal representation of a spreadsheet (thanks to Ric Kamicar for the patch)
- Openoffice: can now read zip-ed files
- Openoffice: can now read files from http://-URL over the net
- excel: removed debugging output
- Openoffice: Exception if an illegal sheet-name is selected
- Openoffice/Excel: no need to set a default_sheet if there is only one in the document
- Excel: can now read zip-ed files
- Excel: can now read files from http://-URL over the net
- Excel-objects: the methods default_sheet= and sheets can now handle names instead of numbers
- tests can now run on the client machine (not only my machine), if there are not public released files involved these tests are skipped
- there was ONE false require-statement which led to misleading error messageswhen this gem was used
- robustness: Exception if no default_sheet was set
- new method reload() implemented
- about 15 % more method documentation
- optimization: huge increase of speed (no need to use fixed borders anymore)
- added the method 'formulas' which gives you all formulas in a spreadsheet
- added the method 'set' which can set cells to a certain value
- added the method 'to_yaml' which can produce output for importing in a (rails) database
- ..row_as_letter methods were nonsense - removed
- @cells_read should be reset if the default_sheet is changed
- error in excel-part: strings are now converted to utf-8 (the parsexcel-gem gave me an error with my test data, which could not converted to .to_s using latin1 encoding)
- fixed bug when default_sheet is changed
- Openoffice: formula support
- Excel: float-numbers were truncated to integer
- Openoffice: two or more consecutive cells with string content failed
- Excel: row method implemented
- more tests
- Openoffice: row method fixed
- ID 11605 Two cols with same value: crash roo (openoffice version only)
- more robust call att Excel#default_sheet= when called with a name
- new method empty?
- refactoring
- bugfix in Excel#celltype
- bugfix (running under windows only) in closing the temp file before removing it
- correct pathname for running with windows
- incorrect dependencies fixed
- support for MS-Excel Spreadsheets
- cells with more than one character, like 'AA' can now be handled
- bugfixes in first/last methods
- new methods first/last row/column
- new method officeversion
- new method row()
- fixed some bugs
- more ways to access a cell
- Initial release