s1kd-flatten
Description
khzae.net
khzae.net
All
Add --xml-catalog parser option.
General
The s1kd-flatten tool combines a publication module and the data modules it references in to a single file for use with a publishing system.
Data modules are by default searched for in the current directory using the data module code, language and/or issue info provided in each reference.
Usage
s1kd-flatten [-d <dir>] [-I <path>] [-cDfimNPpqRruvx] <PM> [<DM>...]
Options
-c, --containers
Flatten referenced container data modules by copying the references inside the container directly in to the publication module. The copied references will also be flattened, unless the -m option is specified.
-D, --remove
Remove unresolved references.
-d, --dir <dir>
Directory to start search in. By default, the current directory is used.
-f, --overwrite
Overwrite input publication module instead of writing to stdout.
-h, -?, --help
Show help/usage message.
-I, --include <path>
Add <path> to the list of directories that the tool will search when resolving references.
-i, --ignore-issue
Always match the latest issue of an object found, regardless of the issue specified in the reference.
-m, --modify
Modify the references in the publication module without flattening them.
-N, --omit-issue
Assume that the files representing the referenced data modules do not include the issue info in their filenames, i.e. they were created using the -N option of the s1kd-new* tools.
-P, --only-pm-refs
Only flatten PM references, leaving DM references alone.
-p, --simple
Instead of the hierarchical PM-based format, use a simpler "flat" format.
-q, --quiet
Quiet mode. Errors are not printed.
-R, --recursively
Recursively flatten referenced publication modules, copying their content in to the "master" publication module.
-r, --recursive
Search directories recursively.
-u, --unique
Remove duplicate references within the PM content.
-v, --verbose
Verbose output. Specify multiple times to increase the verbosity.
-x, --use-xinclude
Use XInclude rather than copying each data module's contents directly inside the publication module. DTD entities in data modules will only be carried over to the final publication when using this option, otherwise they do not carry over when copying the data module.
--version
Show version information.
<DM>...
When using the -p option, the filenames to include can be specified manually as additional arguments instead of searching for them in the current directory. When not using the -p option, additional arguments are ignored.
<PM>
The publication module to flatten.
In addition, the following options allow configuration of the XML parser:
--dtdloadLoad the external DTD.--hugeRemove any internal arbitrary parser limits.--netAllow network access to load external DTD and entities.--noentResolve entities.--parser-errorsEmit errors from parser.--parser-warningsEmit warnings from parser.--xincludeDo XInclude processing.--xml-catalog <file>Use an XML catalog when resolving entities. Multiple catalogs may be loaded by specifying this option multiple times.
Exit status
0
No errors.
1
The publication module specified is malformed.
2
An encoding error occurred.
Example
$ s1kd-flatten -x PMC-EX-12345-00001-00_001-00_EN-CA.XML > Book.xml