Icon Exports Improvements

[csmoore]

Not requirements, just a few suggested changes to the icon exports for when/if this ever gets updated

• Add some columns (to the right of Tags) for Unique ID and Geometry Type
  1. I realize these are already in the tags, but also having in a separate column will make these easier to work with
• It would be helpful if the icon export process did some validation:
  • Check that the Unique ID and Icon Name match as expected
  • Check that the Unique ID and Unique Name are in fact unique
  • Check if the expected svg file actually exists if the user defines an expansion constant/location for {Symbols_Root} in the source code
  • Note: I originally envisioned that the icon export process would check/validate these files when it created the export file & only then add that Note: "image file does not exist;" - right now it just adds that note for everything which doesn't add a lot of value. I understand it probably give this warning because "{Symbols_Root}" doesn't exist but perhaps the export code could expand this to some constant used in the code and do the "exists" check on export.
• Adding the 2525Charlie ID to the tags (already noted here: Include 2525C Symbol ID Code in Export Tags where applicable #86)
• Better Categories for Control Measures, METOCs, e.g. 'Control Measure : Point','Meteorological-Atmospheric : Point' since 'Main Icon' doesn't seem quite right for these

[abouffard]

It actually can check the actual path and only notes really missing image files, IF you configure settings in the jmsml.config file with your actual image information. I have a custom version of jmsml.config on my machine that I use for that purpose, but since its hard coded to my settings its not the version in the repo. I haven't written explicit documentation to explain this yet, but will. It is alluded to in the current jmsml.config itself, if you examine it.

For purposes of documenting it here:

SVGHome = the actual path of your svg files, currently set to point to the relative location of the svgs in the repo, but would need to be changed if the user wanted to move those or point to a different version of the svgs.
GraphicRoot = the string prepended to the paths that are written out in the image file, name, category, tag exports.
GraphicExtension = the file extension used when the export code tests for the existence of your image files, so you can use this to test existence of the svgs or converted emfs. Its the extension written out in the first column of the image file, names, category, tags exports..

<ETLConfig DomainSeparator=" : " PointSize="32" SVGHome ="..\..\..\..\..\svg\MIL_STD_2525D_Symbols" GraphicRoot="{Symbols_Root}" GraphicHome="path_that_equates_to_your_{Symbols_Root}_goes_here" GraphicExtension="emf">

I made the graphic extension variable because the examples I saw had both svg and emf extensions. I don't have emfs on my machine, just the JMSML svgs, so the config file is currently "generic" and needs editing by the user, and the current samples reflect that. That's why the samples (note I think of these as "samples", not real export data, which the user should generate for themselves, IMHO) show all files missing. .

[csmoore]

If it already has this capability & you know how to configure it then maybe when you get an opportunity you update the files at https://github.com/Esri/joint-military-symbology-xml/tree/master/samples/imagefile_name_category_tags

It would be useful if folks could just use these files "as-is" from the repo without needing to configure/compile (which is the primary goal of these suggested improvements). It is a useful output.

I know for me personally I would prefer if the github-posted version notes column didn't have a bunch of spurious "'icon is MAIN'', "image file does not exist" & just had the conditions we needed to pay attention to/check - this could have potentially helped us find some problems earlier.

[csmoore]

The rest of these are just ideas for improvements/testing, but I would like that ""image file does not exist" removed from the example/sample outputs (and only show up there if the expected image file really doesn't exist - so we know this important case) - let me know if I need to open a separate issue for that...

[abouffard]

Fixed in PR #97

[csmoore]

@abouffard I just noticed this was assigned to me - am I supposed to verify? (but it looks like only item 2 or 2iii from the original issue is done - correct?)

[abouffard]

Yes, I just wanted you to be aware that you might want to track these as they got checked off. But of your list of suggestions just the file checking (against the svg files in the repo) have been implemented in the sample exports, because Mike has supplied another set of files and I wanted to make sure nothing was missing. Hang on verifying anything for now though because I am about to pull/merge the newest files, including new dashed frames for planned status support.

[csmoore]

Update: Changed issue to Task List & plz see if this can be included in sprint

[csmoore]

Closing this one, the major things I wanted accomplished from above got done in #108 (the other 2 item are optional and just ideas for future minor improvements)