BugList
Summary Bug List [edit section]
Here is a table of open bugs:
record_id ^ | summary ^ | assigned_to ^ | priority ^ | status ^ |
0015 | TableOfContents doesn't read headings from included page | Tim | low | open |
0013 | Can't edit-in-place the buglist | Tim | high | open |
0012 | ProgressChart macro date parsing error | Tim | low | open |
0031 | sort link sorts all tables on the page, not just the associated table | Tim | low | open |
0020 | Block text missing \n | Tim | low | open |
0021 | 'what links here' doesn't work for some link syntaxes | Tim | medium | open |
0027 | RegressionCheck shows some pages with diffs, even when no diffs are shown on diff page | Tim | medium | open |
0002 | progress chart time parse failure | Tim | low | open |
0004 | Shouldn't save identical page data | Tim | low | open |
0007 | Bug in checkbox field handling | Tim | medium | open |
0026 | Section editing adds a new line to the section | Tim | low | open |
0003 | Can't cancel page create | Tim | low | open |
0029 | Problem with macros and bullets | Tim | medium | open |
0008 | Missing current page in history list | Tim | low | open |
0005 | Exception on invalid search expression | Tim | high | open |
0032 | table edit-in-place removes configuration attributes | Tim | medium | open |
See BugListConf
And BugListLongQuery (not currently used by tables on this page)
Long Form [edit section]
record_id ^ | summary ^ | priority ^ | description ^ | assigned_to ^ | status ^ |
0017 | Add row to attr table uses wrong record_id | <no data> | If you add a row to an attribute table, then the record_id saved in the newly created file is the numeric record_id, instead of the record_id value from the add form.
Note that the attribute filename is correct. To reproduce, add a new record to the User database (see TestUserTable
|
Tim | fixed |
0016 | missing blank line in preformat block | low | See SampleBlock page
Hmmm. There is no SampleBlock page. What was this bug about again?
|
Tim | rejected |
0015 | TableOfContents doesn't read headings from included page | low | The TableOfContents macro doesn't read headings from an included
page. See TestMacroInclude
|
Tim | open |
0014 | Duplicate bullet on big outdent | <no data> | The line of text for a bullet is repeated when a bullet is outdented more than one level.
|
Tim | fixed |
0013 | Can't edit-in-place the buglist | high | Try editing-in-place the Bug List table. You get:
Traceback (most recent call last): File "/var/www/cgi-bin/tbwiki.py", line 47, in main(req) File "/home/tbird/work/tbwiki/cgi-bin/tbwiki_engine.py", line 1529, in main processor_func(req) File "/home/tbird/work/tbwiki/data/plugins/ProcessorTable.py", line 36, in edit_table_in_place message = table.edit_table_in_place(req) File "/home/tbird/work/tbwiki/cgi-bin/table.py", line 1086, in edit_table_in_place tb.show_edit_table_in_place_form() File "/home/tbird/work/tbwiki/cgi-bin/table.py", line 914, in show_edit_table_in_place_form print line_edit_form_spec % form_fields File "/home/tbird/work/tbwiki/cgi-bin/table.py", line 163, in __getitem__ record_id = self.record["record_id"] KeyError: 'record_id' possible solutions: remove 'edit-in-place' link for attrdb databases. But I think this is really just a typo somewhere. There's a fixthis for fixing the schizm between usage of 'record' and 'record_id' on the todo page.
|
Tim | open |
0012 | ProgressChart macro date parsing error | low | ProgressChart macro can't parse date strings
without the time portion of the string
|
Tim | open |
0011 | Need to handle https URLs | <no data> | TBWiki doesn't automatically create link to 'https:' URLs.
|
Tim | fixed |
0010 | RegEx processor doesn't read match lines from processor block | <no data> | RegEx processor doesn't read match lines from processor block
|
Tim | fixed |
0031 | sort link sorts all tables on the page, not just the associated table | low | Go to http://bird.org/tbwtest/TestTableConfig,
and click on the sort link for the "Salary" column.
A new page appears and all the tables have been sorted by Salary. Pages that don't have a Salary column are sorted by some other field, and an error message is displayed. This should be fixed by adding the block_name to the sort_link, so that only that table gets sorted by the indicated field.
|
Tim | open |
0020 | Block text missing \n | low | Block text is missing the \n between the block name
and the block content.
See http://localhost/tbwiki/TestBlocks#named_block
|
Tim | open |
0019 | Merge of database configuration doesn't work as expected | medium | See BugList Long Form, which should have a description in the listing. The cols field in the #!Table processor block should override the cols listing in the BugListConf file | Tim | fixed |
0018 | (DUP of 13) Can't edit-in-place the buglist | <no data> | There's an python exception if you try to edit-in-place the bug list. | Tim | duplicate |
0021 | 'what links here' doesn't work for some link syntaxes | medium | Try "what links here" for the page UnitTestSingleBoldWord.
There are no entries, event though the page List of unit tests has a link to that page.
|
Tim | open |
0027 | RegressionCheck shows some pages with diffs, even when no diffs are shown on diff page | medium | In particular, no diff is shown by RegressionCheck.show_diff, for pages tbwiki.header and TestMarkupVariablesStatic. However, these are shown as errors in the Regression Check table.
I added code to hold messages. It now looks like something related to the variable. I moved this to be dynamically generated by the data_class in tbwiki_engine, but there's still some sort of problem that I can't pin down. There are now lots of debug statements (commented out) in the RegressionCheck main procedure.
|
Tim | open |
0002 | progress chart time parse failure | low | trailing spaces on time entry for progress chart cause a parse failure | Tim | open |
0025 | bug db files have wrong digit format | low | I changed the bug database to use next_page_num for the default_value, but this is always a 4-digit number. bug page made before this change have a different number of digits in
their filenames.
I should go back and fix the filenames to be more sane. The record_id for each file should also be made to match the filename.
|
Tim | fixed |
0004 | Shouldn't save identical page data | low | If you save the identical page, it does a regular save. It should see that contents are the same and avoid the save. | Tim | open |
0023 | single percent page fault | high | if you place a single percent on a page, in some cases you get a page fault (traceback error).
The single percents need to be escaped before the page is formatted with data from the wiki engine. Here is the traceback: Exception raised by tbwiki software Traceback (most recent call last): File "tbwiki.cgi", line 53, in main(req) File "/home2/birdorg/work/tbwiki/cgi-bin/tbwiki_engine.py", line 1513, in main html = block_to_html(req, page_data) File "/home2/birdorg/work/tbwiki/cgi-bin/tbwiki_engine.py", line 1030, in block_to_html new_data = data % req.data ValueError: unsupported format character ' ' (0xd) at index 268 See http://bird.org/test/TestCornerMarkup
|
Tim | fixed |
0006 | Bold and italic processing problems | <no data> | Can't combine bold and italics. Can't apply bold and italics to other elements. | Tim | fixed |
0007 | Bug in checkbox field handling | medium | 6. Checkbox form fields don't work so well
|
Tim | open |
0026 | Section editing adds a new line to the section | low | When you edit a section, an extra newline is added at the end of the section. See either get_section() or set_section(). I think this has to do with the simple '\n'.join() at the end of the routine. | Tim | open |
0001 | this is the first bug | low | This is a multi-line
description of the first bug.
|
Tim | fixed |
0024 | (DUP of 13) Can't edit-table-in-place an attrdb | high | If you try to do edit-table-in-place with an attrdb, it doesn't work. There is a fault with an error about missing record_id.
Duplicate of bug 0013. May be related to bug 0017.
|
Tim | duplicate |
0003 | Can't cancel page create | low | If you press Cancel on a page create, it takes you back to the edit form for the create again. | Tim | open |
0028 | edit comment doesn't show up in history list | low | Add a comment when editing a page, then click on "history".
The comment field is blank. The comment is in the logfile, though. Resolution note: It turned out that the log had entries for both 'edit' and 'get' for the page with the same timestamp, and the 'get' (which has no comment) was being used by the history routine.
|
Tim | fixed |
0029 | Problem with macros and bullets | medium | If you put a macro as the first part of a bullet, then the bullet is not processed correctly.
See http://bird.org/test/TestBullets
|
Tim | open |
0008 | Missing current page in history list | low | History is missing the current page - it can't see the latest diff | Tim | open |
0009 | Page name with space not processed properly | <no data> | Page with space (on TestLinks page) isn't processed properly by makeurl
|
Tim | fixed |
0005 | Exception on invalid search expression | high | You get a python exception if you enter an invalid regular expression in the search form. | Tim | open |
0030 | color coding and types don't work on tbwikidb tables | medium | field color coding, and in general the table field type system doesn't work for tbwikidb databases.
The reason the colors don't work, is that trailing whitespace is left on the field values when they are parsed from the tbwikidb files (/pages). Maybe this could be solved with a simple white-space removal in the table handling during the color compare operation.
|
Tim | fixed |
0022 | datascan module doesn't support multi-line fields | <no data> | The datascan module doesn't support multi-line fields.
Multi-line fields are supposed to be parsable using a field and field_endpat combination, but I can't get it to work. See the function parse_data in data_scan.py, and there is a FIXTHIS saying that this functionality was never tested. See http://bird.org/test/TestTableParsedPages for a test which shows the problems encountered. NOTE: most of this problem is fixed, with changes to data_scan:parse_data(). What remains is to avoid putting the material before the start of the pattern into the value.
|
Tim | fixed |
0032 | table edit-in-place removes configuration attributes | medium | If you choose to edit-in-place a spreadsheet that is an inline moin table, when the table data is written back out, it removes the is_spreadsheet attribute. | Tim | open |
020 | apparently, anyone can enter a bug | <no data> | A weird bug was entered in the bug database on 2015-12-20, not by me. This wiki is supposed to be limited editable only by me. So there must be some permission problem.
See the history for file: bug9HuMcJvz Here is the original description, which I did not write: Yes Infact, they have allowed us to chosoe from a variety of themes for our Gmail inbox too. Also, orkut has fully integrated google talk so that we can use gtalk fully on orkut itself.[:)] ..I'd suggest you to follow the orkut blog too[:)]
|
tim | duplicate |
Outstanding TBWiki bugs (in text format) [edit section]
2. If you save the identical page, it does a regular save. It should see that contents are the same and avoid the save.
3. You get a python exception if you enter an invalid regular expression in the search form.
4. Recent changes is missing newly created pages.
6. Checkbox form fields don't work so well
- checkbox value is not inserted in the edit_record form
- if checkbox is unchecked, the form handling code can't read the value from the form.
11. RegEx processor doesn't read match lines from processor block