[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Thu Jun 9 14:08:10 -0700 2005 --- Moved my vote from issue 48926 . I am beginning to wonder if there is any need to emphasise the HTML element, although that does aid the discovery of the originating localle. I can, however, concieve of situations where the data in the clipboard could have come from sources other than a web browser, and hence 'paste special - as numbers' would be appropriate. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Thu Jun 9 05:41:23 -0700 2005 --- *** Issue 48926 has been marked as a duplicate of this issue. *** - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Tue May 31 12:25:50 -0700 2005 --- adding me to CC: - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Thu May 26 10:34:44 -0700 2005 --- Tha patch fixes issue 38494 , too. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Thu May 26 04:01:55 -0700 2005 --- Søren, I don't know why you think that in a 'de' locale the results of the patched version are ok now. They are for the two cases you marked with a red background, but introduced are three more date values that haven't been a date before, namely IDs 4 (1.2), 6 (1.23) and 18 (1.2345). Furthermore 6 results are now strings that were numbers before, IDs 13 (1,234.5), 21 (1234.5), 22 (123.45), 31 (1.23456), 32 (1,234.56) and 39 (12,345.6). These are the cases why that special handling was introduced: to be able to read en_US number values. The remaining date values are yet another story and the entire date recognition must be straightened out for it. Again: there is no way to automatically determine from a HTML document reliably if 1.234 represents 1234 (group separator) or 1.234 (decimal separator). Eike - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed May 25 16:53:48 -0700 2005 --- Dear Falko, Please find attached patch (attachment 26566) and the tested testcases (attachment 26567). Requesting patch review. Søren - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 User thing changed the following: What|Old value |New value Attachment data| |Created an attachment | |(id=26567) Testcases --- Additional comments from [EMAIL PROTECTED] Wed May 25 16:48:15 -0700 2005 --- Created an attachment (id=26567) Testcases - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 User thing changed the following: What|Old value |New value Attachment is patch| |Created an attachment | |(id=26566) Simple but | |effective patch --- Additional comments from [EMAIL PROTECTED] Wed May 25 16:47:00 -0700 2005 --- Created an attachment (id=26566) Simple but effective patch - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed May 4 12:28:09 -0700 2005 --- On 1.1.3 pt_BR the problem also happens. The issue, however, seems to be related to whether the paste event is sent when the cell is opened or not. If I select the cell and paste the 1.000 value copied from an HTML page, the number 1 is displayed. However, if I double-click the cell and then paste, the correct number, i.e. 1.000 is displayed. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Apr 2 01:42:10 -0800 2005 --- Here are the bugged locales: bs_BA es_CL da_DK es_ES de_DE el_GR es_PE es_UY eu fo_FO fr_BE lt_LT nl_BE nl_NL hr_HR ia id_ID is_IS pt_BR it_IT pt_PT km_KH rw_RW sh_YU sl_SI sr_YU tr_TR uk_UA es_AR ca_ES es_CR de_AT de_LU es_PY gl_ES - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Fri Apr 1 12:32:06 -0800 2005 --- Almost as many in the world are using (,) as thousands separator as thouse using (.) At least 100 million are using (,) That should be reflected in OOo - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Mon Mar 21 07:25:16 -0800 2005 --- I agree that issue 38494 is another derivation of the same underlying technical defect; apparently formatted text (such as HTML) is always pasted in English locale (ie. . is the decimal). For example if I simply copy the text "1.000" (one thousand in German, Danish etc.) from Writer it is recognised as "1" when pasted in Calc. Please raise priority if you agree that this is a serious bug that can cause data loss. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Mar 19 04:41:41 -0800 2005 --- Eike, > Interpreting according to Calc's locale only is most certainly not what > you want, it would misinterpret the great majority of web documents, > being in US-English. And the lang tag isn't really helpful, except maybe > for Danish. What I don't want is Calc's current inconsistent locale guessing strategy that makes pasted HTML untrustworthy. It confuses both existing users and migraters from Excel. According to this 3 year old statistics about 50% of the webpages on the internet is in English. http://www.netz-tipp.de/sprachen.html (the page is in German...) Today, you can't assume that all HTML is english. However, I concur that the lang tag doesn't work wonders. It is just a suggestive input for an enhancement (and I will file it as such elsewhere). SÃren - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 16 05:21:54 -0800 2005 --- Hi Just to sum up: When using locales other than english, Calc converts BOTH â.â and â,â to a decimal separator. It happens not only when copying from html-pages but also from Word, Writer etc. This is a serious bug as it causes data loss. Therefore i suggest changing to prio 2, or at least prio 3. In 2.0 beta, you can specify if the decimal separator key should be âSame as local setting ( , )â but it doesnt work. It still converts â.â to a decimal separator. (Tools > Options > Language settings > Languages) N.B. It seems like Issue 38494 describes the same bug! Thanks! - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Mon Mar 14 03:53:03 -0800 2005 --- In Excel XP and 2003 you can specify whether number recognition should be handled by system settings or you can manually specify the thousands and decimal seperators.( Tools > Options > International tab) (This means that if your locale is danish, an import of 1,000,000 will be interpreted as text and not as the number 100) The good thing about this solution is that it puts the user in control instead of letting Calc try to guess the right format. Could this be a possible solution for Calc? JC - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Mon Mar 14 03:24:48 -0800 2005 --- Soeren, > Excel (locale = da) recognises all 3 tables in the attached testcase as: > 1.000 1.000.000 > 1 1,000,000 Of which 1 is wrong, and 1,000,000 is probably a string, and not a number. > In the short run I suggest that Calc should follow Calc's locale only > to fix the imminent defect. Further up the road HTML's lang tag could > define the locale - if the lang tag is set for either the table or the > whole page. The latter step can be regarded as an enhancement. Please read again what I wrote previously on Tue Mar 8 13:08:23 -0800 2005 Interpreting according to Calc's locale only is most certainly not what you want, it would misinterpret the great majority of web documents, being in US-English. And the lang tag isn't really helpful, except maybe for Danish. Eike - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Mar 12 06:33:59 -0800 2005 --- I have opened issue 44872 "Link to External data can't handle XHTML 1.0 Strict tables": http://www.openoffice.org/issues/show_bug.cgi?id=44872 - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Mar 12 04:36:32 -0800 2005 --- Excel adheres the locale setting of the spreadsheet exclusively. As far as I can tell it doesn't evaluate the locale of the source document. Quattro Pro 12 acts the same. In the short run I suggest that Calc should follow Calc's locale only to fix the imminent defect. Further up the road HTML's lang tag could define the locale - if the lang tag is set for either the table or the whole page. The latter step can be regarded as an enhancement. SÃren PS. As Claus points out there is a problem with opening the XHTML testcase, but that is another issue. Is there an issue filed for this problem? - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Mar 12 03:19:42 -0800 2005 --- The first example in XHTML 1.0 Strict works the same way as the second example in HTML 4.01 Transitional when opening in Calc. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Mar 12 03:13:32 -0800 2005 --- In 680m82 in Danish I can't insert data from hyperlink with the first example using XHTML 1.0 Strict from thing. But the second using HTML 4.01 Transitional gives these results: Table without lang tag 1 100 should have been 1000 100 1 1,000,000 (textstring) should have been 1 1,000,000 (textstring) Table with lang = da 1 100 should have been 1000 100 1 1,000,000 (textstring) should have been 1 1,000,000 (textstring) Table with lang = en 1 100 should have been 1 1.000.000 (textstring) 1 1,000,000 (textstring) should have been 1000 100 - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Sat Mar 12 00:05:06 -0800 2005 --- Excel (locale = da) recognises all 3 tables in the attached testcase as: 1.000 1.000.000 1 1,000,000 Changing the lang tag for the entire page doesn't change Excels number recognition. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 User thing changed the following: What|Old value |New value Attachment data| |Created an attachment | |(id=23718) Testcase (HTML | |4.01) --- Additional comments from [EMAIL PROTECTED] Fri Mar 11 23:57:39 -0800 2005 --- Created an attachment (id=23718) Testcase (HTML 4.01) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 User thing changed the following: What|Old value |New value Attachment data| |Created an attachment | |(id=23717) Testcase --- Additional comments from [EMAIL PROTECTED] Fri Mar 11 23:33:18 -0800 2005 --- Created an attachment (id=23717) Testcase - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 9 05:05:38 -0800 2005 --- Then you could also select the numbering format as an option when importing HTML to Calc or External Data from the Insert menu in Calc. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 9 04:04:48 -0800 2005 --- Jens, > Spreadsheets like Excel and QuattroPro has a dialog box in which the user can > set numberingformats on each HTML import. Don't know about QuattroPro, but I've never seen it in Excel. Where did you encounter it? Eike - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 9 03:57:53 -0800 2005 --- Please bear in mind that the issue not only arises when inserting from the clipboard but also when opening files, including http/ftp, via the Calc HTML filter. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 9 03:27:54 -0800 2005 --- Even though you are using OpenOffice.org in Danish you will often get tables from URL's in English. I would suggest that you can select the number format when using "Insert special" and not just by changing the locale setting. And when using the normal "Insert" then use the locale setting. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 9 02:15:31 -0800 2005 --- Hi Eike Thank you for your analysis. We have discissed the issue and has come to the conclusion that the primary issue is to have the user decide which numberingformat is to be used. The easiest way to do this is to have the OOo locale setting to determine the numberingformat of the HTML import. If the user wants US numebring format of the import the locale should be set to US. If the user wants Danish og German the locale setting of OOo should be set to Danish or German. The current situation is completely confusing since a HTML import with Danish locale setting uses a mix of US and Danish nummering. Spreadsheets like Excel and QuattroPro has a dialog box in which the user can set numberingformats on each HTML import. Thanks JensK - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Wed Mar 9 01:57:04 -0800 2005 --- Hi Eike, Thanks for your in depth analysis. I agree overall and have a few comments. Eike wrote: > Furthermore, pages in English tend to contain no language tag, because "the > language of the web is English", implying an en_US locale. I meant specifying lang="da" for the table - not necessarily the whole page. Either way, if Calc would respect the locale set in HTML's lang tag - only if specified - the user always has a way of dictating the locale. As of now the locale is sometimes lost in translation in the mix between 1 and 2 without any way of dictating the locale. That is why I suggest respecting HTML's lang tag - if specified. W3 spec: http://www.w3.org/TR/REC-html40/struct/dirlang.html I will try to analyse how other office suites handle the issue in general. SÃren - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 --- Additional comments from [EMAIL PROTECTED] Tue Mar 8 13:08:23 -0800 2005 --- I'll chime in for some comments: Soeren wrote: > The web is multilingual nowadays. Pasted HTML should dictate locale if > the pasted HTML contains the lang tag. Otherwise Calc should set the > locale. I'd say that HTML documents in 99.8% of all cases don't specify a locale, in fact I haven't seen any so far. If at all, they will contain the language, which in case of, for example, "en", "fr" or "es" is equivalent to no locale. Only for languages where there is not more than one possible locale it would be sufficient. Furthermore, pages in English tend to contain no language tag, because "the language of the web is English", implying an en_US locale. The majority of all these cases interpreted in the current Calc locale would lead to wrong results if separators didn't match. > To improve number recognition Calc can guess the locale of the pasted > HTML if any value in the the pasted content contains either two > periods or two commas with three digits in between (e.g. 1.000.000). There are locales that don't use dots or commas as separators. There are also locales that use the group separator (AKA thousands separator) not in groups of three digits each, but in mixed forms; currently not supported by OOo, but possible. Hindi for example, AFAIR. The only thing one can say for sure is that a separator appearing more than once can't be a decimal separator, and more than two different separators don't form a number. The way to go for each string would be 0. in case the document specifies a locale, use that 1. look if it could be parsed as an en_US number 2. look if it could be parsed as a number according to the current locale 3. maybe try to detect another meaningful combination of separators Calc currently tries a somewhat weaker method of 1. and 2. for each string and if both apply, en_US is taken. This is why 1.000 in de_DE locale results in 1. The checks, however, could be refined to stick to one locale once detected, and better take the group separator into account. Still, if there are several numbers like 1.000 and 2,000 you can't say which is what. This is the amount of uncertainty a detection could not solve, like it is the case for http://investdb.theglobeandmail.com/invest/investSQL/gx.price_history?pi_symbol=UALAQ-Q2 and only a dialog would help. > Furthermore, if digits are separated by both a period and a comma it > is almost certain that the last of these is the decimal. (e.g. > 1.000,01). Seconded. > It is a pretty sure shot that the rest of the contents use the same > thousands separator. Which effectively means parsing the document content twice, at least up to the point where it is certain that both separators were detected and previous assumptions didn't match. Ok, not a big deal, but maybe not nice regarding performance for large documents. Eike - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[sc-issues] [Issue 39898] Number recognition formats pasted value from HTML page of e.g. 1.000 as 1 despite . (period) being thousands separator according to Calc's locale setting
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=39898 User thing changed the following: What|Old value |New value Summary|Number recognition formats|Number recognition formats |pasted value from HTML|pasted value from HTML |page of e.g. 1.000 as 1 |page of e.g. 1.000 as 1 |despite . being thousands |despite . (period) being |seperator according to|thousands separator |locale|according to Calc's locale | |setting - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
