On Wednesday, 01/02/2017 12:20 GMT, Stuart Henderson <s...@spacehopper.org> wrote:
> Another option would be to ask whoever wrote these patches if they can > publish it as a repository somewhere.. It is effectively forked, so > it would be a lot simpler to handle this as an update rather than a > patchset (especially with the regenerated bison parser). Nice to know, so I think that we can follow two routes: * first use debian git repository [1] that contains and maintains all new stuff and apply only patches (debian/patches/) that we want. * second based on the answer of Martin Bayer (html2text maintainer) that wrote to me: << Let me make a long story short: You are right with most of your points and it is true I haven't released a new version for years. But the reason for this is, that html2text relies for its parser on a bison++ version from the early 1990s. In order to go on with html2text, one would need to move to the most recent version of GNU Bison, or re-write the parser, which means to become independent from a parser generator. html2text's unique feature was (at the time) rendering of tables. I understand that w3m does this job too, as links/links2 do also. So it's hard for me to see a good reason to invest time and effort in html2text. As I'm focused for my work now to themes outside the open source community, I'm more likely going to discontinue html2text rather than making these efforts. >> remove html2text and force users to use w3m o links/links2. For example you can convert html to text with w3m with the following command: $ w3m -dump file.html > text what do you think? /davide [1] https://anonscm.debian.org/cgit/collab-maint/html2text.git/ -- PGP: 0x90e516091925019b | 72ED 84B1 0123 8744 760B AB89 90E5 1609 1925 019B
