torvalds / uemacs Goto Github PK

In dolock, shouldn't the declaration be:
lname[MAXNAME], locker[MAXLOCK + 1]...

Instead of:
lname[MAXLOCK], locker[MAXNAME + 1]...

This causes the function to read an arbitrary 128 bytes from the file instead if 512, which I don't understand why.

I did a quick recheck after last commit setting up NFILEN to 256

• em crashes when called with filename argument longer than 255 [easy to fix with mystrscpy instead of strcpy]

• as em variables size are limited to 127+1, longer filenames will create problems with macroes [ex: $cfname holds the filename of current buffer]

• write-file command can be called with strings up to and longer than 255, so em will create files having up to 255 characters, but can't open them either with find-file command or as argument from command line [LOCK ERROR: cannot access lock file - File name too long]

• filenames completion works, a manual display refresh [^L] is necessary to clean up the display.

attached is a simple em script I used to generate the files for test: [strip .txt extension]
tstfnam.cmd.txt

I would suggest to limit NFILEN to 128 for now and fix the strcpy of filename arg issue.

that is the question

@torvalds thank you for inspiring the Linux movement. I would place this issue in a better place if I knew where to; This question is simple, "What gives open source the right to take away root user?" termux/termux-app#507 (comment) specifically in regard to the Android operating system. Is Android the most popular derivative of Linux today with billions of users? Probably.

One of many time and resource consuming examples: the way Android selects and corrects text to be recorrected and reselect repeatedly by the user is inline with no root user login to Android rational.

Cheers ☕ 🎸

PS em builds beautifully in an Android smartphone through Arch Linux in Termux PRoot with warnings:

this editor does not support Chinese very well...

I have been working on a fork since 2013, so believe me there is plenty to fix. :)

I may be mistaken, but I think the function unicode_to_utf8 is broken. Although it is able to generate the continuation bytes of the UTF8 sequence, it fails to produce the correct leading byte.

Please forgive my impertinence, I'm a MicroEMACS noob - didn't start using it until the fall of 1986.

I'm running 4.0.15 on various Linux distros and my work involves a lot of csv files that are getting awk/grep/sed treatment. When I save a file I've edited with MicroEMACS I often find a single newline appended, even if I went nowhere near the end of the file. If I jump to the end, remove the newline, and leave the cursor at the end of the final line with text then I get the expected result.

This should not be difficult to reproduce, I encounter it several times a day. I do not think it's universal, but I can not yet describe the sort of editing that triggers this event.

linus Please tell me how to configure and enable line numbers in uemacs

Sample Content

Who merges pull request in this repository? is linus? or others?

Hello @torvalds ,
I search most of the place but i am unable to find out the services of VRF used in linux.
Can you please tell me the services of VRF and also the methods so I can integrate this VRF services in Linux.

Thanks

Hope to see you better

or am I missing something?

🦄

When I press C-s or C-x C-s, uemacs stops working, I tried to recompile, but it doesn’t help me (I didn’t change anything in the code), it freezes anyway. Does anyone have the same problem? Has anyone found a solution?