Hm hm, good evening! First, Johannes Schöpfer is not subscribed to this list – i had forgotten that this afternoon, but in the meantime i have forwarded your message to him.
Steffen Nurpmeso wrote in <20181129133635.guz8p%[email protected]>: |[email protected] wrote in <[email protected]>: || Quoth Herr Steffen Nurpmeso, 'We are at Slackware, too! Thank Oh, Herr! Normally "Junger Herr" ("Young Mr."). But thanks :) You can use the command escape "~Q MESSAGE LIST" in compose mode to quote messages, by the way. We have gained some new command escapes in general, and some flexibility regarding the look and feel of quoted messages, too. If you are using web browsers you could look at or download the fully cross-referenced manual on/from my web site[1], in case of interest it might even make sense to keep it as a reference, a cheat sheet. I could not live without the internal cross-references of the manual myself .. though i use it from within the TTY, from a patched less(1). But that is a different story. [1] https://www.sdaoden.eu/code-nail.html ||you, this is really nice!! I did not even know that Slackware is ||still truly alive, but i think: good that it is.' || Viva Slackware! || || Quoth Herr Johannes Sch�pfer, 'Maybe a missing user folder in ||/var/spool/mail ?' || No. || || Quoth Herr Johannes Sch�pfer, 'Where does the string "mailx ||v14.9.11, 2018-08-08 (build for linux)" come from? Is this a prebuilt ||binary? Or from slackbuilds.org, or from slackware-current?' || I thought I got it from 'mail -V', but it is the the first ||line of 'mailx -Xversion'. This is in slackware-current. I've been Maybe we could extend -V if *verbose* a.k.a. -v was given before, too; on the other hand, this output is already incompatible with what was there in the past, i will change this, thanks. ||using mail since the beginning, even as an MUA (!!!!!). Which means I ||am soon headed for the grave, so maybe you don't have to address my ||problems. I do what i can. That includes your (and mine) MUA experience, if possible. || Quoth Herr Johannes Sch�pfer, 'heirloom-mailx development is ||now at s-nail, so why not go with it.' || That's the spirit! || || nytimes is a plain text file. I save e-mail into files named ||after the sender (for example JohnSmith, SteffenNurpmeso, ||JohannesSch�pfer, nytimes...). || I just noticed something: || || s2 NYTimes || ||works; || || s2 nytimes || ||returns the error. nytimes is a link to NYTimes. | |Ah! Oh yes, i see, this is true. |This is because we actively reject to follow symbolic links by |setting O_NOFOLLOW when calling open(2), and our error number |mapping table misses the ELOOP error that is generated if this is |triggered! This i will fix for the next release, hopefully before |the christian christmas -- thanks for reporting this bug, Russell! Yes, indeed i am totally overworked and did not realize that what could be seen was not the [E]PIPE error, but that it was simply _the_ indication that we did not support [E]LOOP yet: #?0[steffen@essex nail.git]$ mail \ -X'echo $^ERRDOC-PIPE and $^ERRDOC-LOOP' -Xx Broken pipe and Snorkeling on empty pipe I also realized that the behaviour regarding symbolic links is mentioned in the "A starter" manual section, which made me wonder whether you were actually asking for a switch to turn this behaviour off? The problem with the codebase currently is that it mingles things: i think it would be more or less ok to follow symbolic links _unless_ the target had to be created anew, but this will be hard to implement now, it anyway will require time. But some variable to follow symbolic links would be doable; what do you think? Anyway, the missing LOOP error is now on the [master], and all [stable/] branches, including the credit due to you! Thanks for reporting this bug, and hopefully we just work for you! Ciao, and a good evening from Germany. --steffen | |Der Kragenbaer, The moon bear, |der holt sich munter he cheerfully and one by one |einen nach dem anderen runter wa.ks himself off |(By Robert Gernhardt)
