You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently Pandoc supports looking for data files in the directories
$DATADIR/filters
$DATADIR/defaults
$DATADIR/metadata
$DATADIR/templates
$DATADIR/csl
but the following potentially equally useful data subdirectories are not currently supported:
$DATADIR/reference-docs
$DATADIR/abbreviations
$DATADIR/highlight†
$DATADIR/includes‡
$DATADIR/bibliographies
Notes to the above list:
†: for both highlighting styles and syntax definitions
‡: for both header and body includes
so I wonder if there are any reasons not to support any or all of these, and similarly for other kinds of data files which may be added in the future and/or I have missed above.
Some reasons why this may be a good idea:
Consistency with those already supported.
Otherwise there is a strong incentive to just drop everything into the root of $DATADIR which is messy and hard to maintain.
Ease of (local) deployment:
You can have people put everything in the appropriate subdirectories (or put them there for them)1
People can then use simple file names on the command line, without having to remember/be told that they have to use a full path in some cases but not in others. (This last is an issue even if you have a predefined environment variable PANDOC_DATADIR vel sim. or use ${USERDATA} in a defaults file!)
Having separate directories makes it easier to use a consistent naming scheme for example for a filter and a related reference doc.
Alternative solutions
Tell people/train yourself to consistently using a predefined prefix environment variable like the above-mentioned PANDOC_DATADIR, and ${USERDATA} in defaults files. This shouldn’t really be necessary.
I hope implementing this for all kinds of data files makes it easier rather than harder to implement!
Unfortunately I can't see myself learning Haskell, so I depend on people's good will to implement it.
Footnotes
In most (all?) modern GUIs you can just drop a directory containing properly named subdirectories on the user data directory and they will be automatically merged. ↩
The text was updated successfully, but these errors were encountered:
Describe the problem
Currently Pandoc supports looking for data files in the directories
$DATADIR/filters
$DATADIR/defaults
$DATADIR/metadata
$DATADIR/templates
$DATADIR/csl
but the following potentially equally useful data subdirectories are not currently supported:
$DATADIR/reference-docs
$DATADIR/abbreviations
$DATADIR/highlight
†$DATADIR/includes
‡$DATADIR/bibliographies
Notes to the above list:
†: for both highlighting styles and syntax definitions
‡: for both header and body includes
so I wonder if there are any reasons not to support any or all of these, and similarly for other kinds of data files which may be added in the future and/or I have missed above.
Some reasons why this may be a good idea:
$DATADIR
which is messy and hard to maintain.PANDOC_DATADIR
vel sim. or use${USERDATA}
in a defaults file!)Alternative solutions
Tell people/train yourself to consistently using a predefined prefix environment variable like the above-mentioned
PANDOC_DATADIR
, and${USERDATA}
in defaults files. This shouldn’t really be necessary.Related
P.S.
I hope implementing this for all kinds of data files makes it easier rather than harder to implement!
Unfortunately I can't see myself learning Haskell, so I depend on people's good will to implement it.
Footnotes
In most (all?) modern GUIs you can just drop a directory containing properly named subdirectories on the user data directory and they will be automatically merged. ↩
The text was updated successfully, but these errors were encountered: