Whole procedure for getting them follows these steps:
- get "readme.txt" content (doesn't matter how deep this file is),
- if there is no such file get some TXT file info (no matter how deep, 1st found IMHO).
It might not always give expected results for the user or can mislead him/her. I was completely surprised what I saw when I've downloaded some C# project from Internet. The content which was shown came back from licence.txt file buried deep inside whole structure.
This might be really nice to give the user two additional parameters for better archives "comments handling".
1. ArchiveCommentDeep
Could specify how deep in folders structure TC should search for any comments. IMHO most of the users will be happy with 1-2 levels inside.
I am not sure if this speeds up whole process but comment's file path will be shorten for sure.
2. ArchiveCommentsNames
Basically only "readme.txt" is taken as search result. It might be even better if we can decide comments files naming conventions in the way:
Code: Select all
ArchiveCommentsNames=readme.txt info.diz read.me czytaj.toIt is worth to mention that #1 should be more important than #2.
Giving additional parameters for files naming could be a bingo for whole idea. Just imagine that TC might look i.e. for the "*_info.txt" files or "<exename>.txt" taking as the result first match.
