I have been playing around with svg reports, especially the relgraph output. I transferred one of the reports to my web-server as an experiment and found that the “person-nodes” which otherwise would have a .png showing a portrait on them are all showing a null graphic. But only on the web server!
The same file shows the proper .png graphics on my development machine so I used a text editor to take a look and it seems that the person-node graphics are not converted to a gramps path in the same way they would be in a Narweb report. In a Narweb report such .png files and .jpeg files are neatly converted to a carefully mapped-out hashed format which is used to assemble them back into html web pages.
It is my observation that this hashing process, done by Gramps, which embeds a graphic within a html document does not occur in the same way with .svg files. Leaving them with ‘orphan’ references when moved off the source machine. However, the same graphics seem to be readily available as pathed references in the output Narweb file structure.
It would seem that the same process which creates the container html files for Narweb graphics could be used for the .svg files (perhaps others too), making them transportable, providing that the web site was resolvable.
Is there something we might might be to do here, with relatively little effort by a gifted, generous, handsome, devs to make our .svg files something we can email our relatives as a front-end to our back-end Narweb sites? I already do this but would like my .svg files to be a little more cosmetic with resolvable graphics files.
I hope this stimulates a bit of discussion
best regards and stay safe!