csl processor bug - second field align
See here:
http://forums.zotero.org/discussion/18669/please-post-update-to-advances-in-complex-systems-style-to-work-with-csl-1/#Item_7
second field align does not work correctly when the search key includes a date.
Examples - IEEE works correctly:
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<layout suffix=".">
The following two also work fine:
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<sort>
<key variable="issued"/>
</sort>
<layout suffix=".">
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<sort>
<key macro="author"/>
</sort>
<layout suffix=".">
But this completely messes up formatting
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<sort>
<key macro="issued"/>
</sort>
<layout suffix=".">
I can consistently replicate that in different styles with second-field-align="flush" and a macro that includes a date as a sort key (doesn't have to be the first sort key, either).
http://forums.zotero.org/discussion/18669/please-post-update-to-advances-in-complex-systems-style-to-work-with-csl-1/#Item_7
second field align does not work correctly when the search key includes a date.
Examples - IEEE works correctly:
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<layout suffix=".">
The following two also work fine:
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<sort>
<key variable="issued"/>
</sort>
<layout suffix=".">
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<sort>
<key macro="author"/>
</sort>
<layout suffix=".">
But this completely messes up formatting
<bibliography et-al-min="7" et-al-use-first="1" entry-spacing="0" second-field-align="flush">
<sort>
<key macro="issued"/>
</sort>
<layout suffix=".">
I can consistently replicate that in different styles with second-field-align="flush" and a macro that includes a date as a sort key (doesn't have to be the first sort key, either).
Some special magic is invoked on macros that contain a date -- the processor extracts the date and uses that for the key, and leaves out anything else contained in the macro. That's unavoidable because date sort-keys are very different from the string representation (the key needs to provide a sorting hint on negative dates, and account for ranged date values).
Something inside the code that unpacks the macro is apparently doing a bad thing. I'll look into it today.