Emacs lsp-mode multiple completions

Recently I started using lsp-mode with company-mode for completions (Following instructions from How to setup Emacs LSP Mode for Go). I was pretty happy with my configuration which for pretty basic using use-package But, soon ran into the following issue where 2 completion minibuffers started showing up.

So, did a bit of googling to land in Multiple completion modes enabled. How to fix this?. It did not fix my issue but, pointed me in the right direction. So, I looked at the current values for company-frontends & company-backends in the buffer in question and found that auto-complete-mode was enabled which was the contributor to the lower completion buffer which I wanted to get rid off. I also remembered disabling global-auto-complete-mode But, the devil is in the details taking a peek into ac-config-default showed the culprit (all the way below in line #10). So, with a little tinkering later we get And, all is well in Emacs :)

Generate file digests using OpenSSL

A simple way to generate file digests in windows which does not have md5sum, sha1sum etc. is to use OpenSSL if it is installed. Here is how you do it

PDFs in the linux commandline

I was recently faced with the task of scanning a lot of documents and wanted to preserve them as multi-page pdfs. Converting the raw jpegs to pdfs proved to be quite easy using ImageMagick's convert that I got from http://stackoverflow.com/questions/8955425/how-can-i-convert-a-series-of-images-to-a-pdf-from-the-command-line-on-linux. But, the file sizes of the pdfs was really large when using convert to create multi-page documents. A little bit of googling and I found pdfunite which was perfect and I was able to create reasonably sized multi-page pdfs from a set of single page pdfs :)

Naruto downloader from managreader.net

After reading Naruto upto chapter 520 (courtesy naruto with elisp) I was eager to read the rest. As of today the latest chapter is 570. I found mangareader.net after a bit of googling and was busy reading my way through the chapters. But, like before reading in the browser was not up to my taste. Currently the mcomix is the comic reader of my choice. So, I set about to write a script to automatically download the remaining chapters from mangareader.net. I do not know if its wrong to do so, the site does not have any terms of use :| So, I let it rip and now I'm busy reading... :)

Project Euler Problem 1 functionally

I'm trying to learn a bit of elisp (see Luhns Algorithm in elisp). So, to try out me elisp some more I tried the Project Euler's simplest problem, Problem 1. So, I cracked open emacs and tried out a few things, gave up and went into hibernation. A few weeks later I remembered this problem and had a go at it again. I thought I nailed it... After a bit of googling I found out the elisp only supports 300 recursive calls :(. So, I decided to take the plunge into clisp and started downloading that. Meanwhile I had some old version of scala available and tried this Wow, that went well and by the time I had googled to find the syntax for scala etc., clisp was downloaded and here goes https://gist.github.com/fc-unleashed/86ecd9aee76dcd8ec1decb3f15b84d6d

Naruto with elisp

I just downloaded a huge Naruto (Manga) torrent which was organized like below

|- Chapter 001 to 010
|---- c001-p01.jpg
|---- c001-p02.jpg
|---- ...
|---- c010-p21.jpg
|---- c010-p22.jpg
|- Chapter 011 to 020
|- ...
|- Chapter 511 to 520
|- ...

I prefer to read my comics in cdisplay/comix which can read cbz/cbr formats. I was looking for an easy way of automating the process in elisp (buoyed by my previous success with the Luhnz algorithm). I was also itching to use this. My attempt at creating a script in one go was a failure and since I wanted to do it quick I used this

M-: (dotimes (i 52) (insert (format "cd \"Chapter %03d to %03d\"\n(dotimes (i 10) (insert (format \"\\\"C:\\\\Program Files\\\\7-Zip\\\\7z.exe\\\" a -tzip \\\"E:\\\\comics\\\\Naruto\\\\out\\\\%s.cbz\\\" c%s-*\\n\" (+ %d i) (+ %d i))))\ncd ..\n" (1+ (* i 10)) (* (+ i 10)) "%03d" "%03d" (1+ (* i 10)) (1+ (* i 10)))))

On running which I got this

cd "Chapter 001 to 010"
(dotimes (i 10) (insert (format "\"C:\\Program Files\\7-Zip\\7z.exe\" a -tzip \"E:\\comics\\Naruto\\out\\%03d.cbz\" c%03d-*\n" (+ 1 i) (+ 1 i))))
cd ..
cd "Chapter 011 to 011"
(dotimes (i 10) (insert (format "\"C:\\Program Files\\7-Zip\\7z.exe\" a -tzip \"E:\\comics\\Naruto\\out\\%03d.cbz\" c%03d-*\n" (+ 11 i) (+ 11 i))))
cd ..
cd "Chapter 021 to 012"
(dotimes (i 10) (insert (format "\"C:\\Program Files\\7-Zip\\7z.exe\" a -tzip \"E:\\comics\\Naruto\\out\\%03d.cbz\" c%03d-*\n" (+ 21 i) (+ 21 i))))
cd ..

Then positioning the caret at the first character for the dotimes line I recorded the following macro C-x ( C-k M-: C-y C-j C-k C-n C-n C-x ). This done, I ran the macro C-x e and held down the e till all occurrences were covered and I got this.

cd "Chapter 001 to 010"
"C:\Program Files\7-Zip\7z.exe" a -tzip "E:\comics\Naruto\out\001.cbz" c001-*
"C:\Program Files\7-Zip\7z.exe" a -tzip "E:\comics\Naruto\out\002.cbz" c002-*
...
"C:\Program Files\7-Zip\7z.exe" a -tzip "E:\comics\Naruto\out\010.cbz" c010-*
cd ..
...

I saved it as a bat file and let it loose and I had a neat little directory in which all the cbz files :). I'm sure there are more cleverer ways to accomplish this but, this was the easiest approach that I could think of (after a few failed attempts at nested inserts and let).

Clean sysouts from java project

Recently I was working in a project that had a lot of sysouts (System.out.println) peppered all over the code. I tried the usual sed approach find . -name '*.java' | xargs sed -i '/System.out.println/d' and that din't work out too well because of multiline sysouts. So, I decided to try the AST approach using the java parser from http://code.google.com/p/javaparser/. So, create a maven project add the javaparser repository and dependency in the pom.xml as given below

And the java code that creates the AST and looks for the System.out.println method call

Luhn algorithm in elisp

I was looking for the checksum validation of credit card numbers and came across the Luhn Algorithm. I had emacs open and on a whim decided I'll try and implement it using elisp. I'm no lisp programmer but, have in the past managed to write some basic elisp in .emacs so, I guessed it would take me about .5hr at most.

I guessed wrong. It took me a lot longer to wrap my head around even some of the simple elisp constructs like let and lambda took quite a while and it took a lot longer than I anticipated. Here, I present to you the fruit of my labors :)