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

	(use-package lsp-mode
	:ensure t
	:config
	(setq lsp-log-io nil) ; only enable for debugging
	(add-hook 'go-mode-hook #'lsp-deferred)
	)
	(use-package lsp-ui
	:ensure t)
	(use-package company
	:ensure t
	:config
	(setq company-idle-delay 0)
	(setq company-minimum-prefix-length 1))

view raw lsp-company-config.el hosted with ❤ by GitHub

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

	(use-package auto-complete
	:ensure t
	:init
	(ac-config-default)
	; (global-auto-complete-mode t)
	)

view raw auto-complete-config.el hosted with ❤ by GitHub

But, the devil is in the details taking a peek into ac-config-default showed the culprit (all the way below in line #10).

	;;;###autoload
	(defun ac-config-default ()
	"No documentation."
	(setq-default ac-sources '(ac-source-abbrev ac-source-dictionary ac-source-words-in-same-mode-buffers))
	(add-hook 'emacs-lisp-mode-hook 'ac-emacs-lisp-mode-setup)
	(add-hook 'c-mode-common-hook 'ac-cc-mode-setup)
	(add-hook 'ruby-mode-hook 'ac-ruby-mode-setup)
	(add-hook 'css-mode-hook 'ac-css-mode-setup)
	(add-hook 'auto-complete-mode-hook 'ac-common-setup)
	(global-auto-complete-mode t))

view raw auto-complete-config.el hosted with ❤ by GitHub

So, with a little tinkering later we get

	(defun fc/lsp-deferred ()
	"When enabling lsp-mode disable auto-complete"
	(auto-complete-mode -1)
	(lsp-deferred))
	(use-package lsp-mode
	:ensure t
	:config
	(setq lsp-log-io nil) ; only enable for debugging
	(add-hook 'go-mode-hook #'fc/lsp-deferred)
	)

view raw lsp-deferred.el hosted with ❤ by GitHub

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

	openssl dgst -md5 <filename>
	openssl dgst -sha1 <filename>
	openssl dgst -sha256 <filename>

view raw openssl-digest.sh hosted with ❤ by GitHub

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 :|

	import re
	from urllib2 import urlopen
	from zipfile import ZipFile, ZIP_DEFLATED
	from xml.dom.minidom import parseString
	def get_info(line, alt_regex):
	try:
	line = line[:line.index('</a>')] + '</a>'
	line = line[line.index('<a href'):]
	dom = parseString(line)
	info = {}
	a = dom.getElementsByTagName('a')[0]
	info['next'] = a.getAttribute('href')
	img = a.getElementsByTagName('img')[0]
	info['img_url'] = img.getAttribute('src')
	info['img_ext'] = info['img_url'][info['img_url'].rindex('.') + 1:]
	alt = img.getAttribute('alt')
	m = re.search(alt_regex, alt)
	info['chapter'] = int(m.group(1))
	info['page'] = int(m.group(2))
	dom.unlink()
	return info
	except Exception as e:
	print('[ERROR] %s' % line)
	print('[ERROR] ' + e)
	def get_image(url):
	try:
	f = urlopen(url)
	b = f.read()
	f.close()
	return b
	except Exception as e:
	print('[ERROR] ' + e)
	def get_chapter(url_prefix, url_suffix, title, chapter=1):
	need_more = True
	alt_regex = re.compile(r'%s (\d+) - Page (\d+)' % title)
	cbz = ZipFile('%03d.cbz' % chapter, "w", ZIP_DEFLATED)
	url = '%s%s' % (url_prefix, url_suffix)
	try:
	while ( need_more ):
	f = urlopen(url)
	lines = f.readlines()
	f.close()
	line = filter(lambda x: x.find('id="img"') != -1, lines)[0]
	info = get_info(line, alt_regex)
	need_more = info['chapter'] == chapter
	if ( need_more ):
	cbz.writestr('%02d.%s' % (info['page'], info['img_ext']), get_image(info['img_url']))
	url = '%s%s' % (url_prefix, info['next'])
	else:
	# new chapter
	cbz.close()
	chapter = info['chapter']
	need_more = True
	cbz = ZipFile('%03d.cbz' % chapter, "w", ZIP_DEFLATED)
	except IndexError:
	pass # image not found so end of chapter
	get_chapter('http://www.mangareader.net', '/naruto/521', 'Naruto')

view raw mangareader.py hosted with ❤ by GitHub

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...

	(defun num_check (x) (if (or (eq (mod x 3) 0) (eq (mod x 5) 0)) x 0))
	(defun sum(x) (if (eq x 3) 3 (+ (num_check x) (sum (- x 1)))))
	(sum 3)
	3
	(sum 5)
	8
	(sum 10)
	33
	(sum 1000)
	Debugger entered--Lisp error: (error "Lisp nesting exceeds `max-lisp-eval-depth'")

view raw luhn.el hosted with ❤ by GitHub

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

	Welcome to Scala version 2.9.0.1 (Java HotSpot(TM) Server VM, Java 1.6.0_20).
	Type in expressions to have them evaluated.
	Type :help for more information.
	scala> def num_check(n: Int):Int = if ( (n % 3) == 0 || (n % 5) == 0 ) n else 0
	num_check: (n: Int)Int
	scala> def do_sum(x: Int):Int = if ( x == 3 ) 3 else num_check(x) + do_sum(x - 1)
	do_sum: (x: Int)Int
	scala> do_sum(10)
	res0: Int = 33
	scala> do_sum(100)
	res1: Int = 2418
	scala> do_sum(1000)
	res2: Int = 234168

view raw luhn.scala hosted with ❤ by GitHub

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

	<project xmlns="http://maven.apache.org/POM/4.0.0"
	xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
	xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
	<modelVersion>4.0.0</modelVersion>
	<groupId>org.fc</groupId>
	<artifactId>sysout-cleanup</artifactId>
	<version>1.0</version>
	<repositories>
	<repository>
	<id>javaparser</id>
	<name>JavaParser Repository</name>
	<url>http://javaparser.googlecode.com/svn/maven2</url>
	<snapshots>
	<enabled>false</enabled>
	</snapshots>
	</repository>
	</repositories>
	<properties>
	<commons-io-version>2.0.1</commons-io-version>
	<java-parser-version>1.0.8</java-parser-version>
	</properties>
	<dependencies>
	<dependency>
	<groupId>commons-io</groupId>
	<artifactId>commons-io</artifactId>
	<version>${commons-io-version}</version>
	</dependency>
	<dependency>
	<groupId>com.google.code.javaparser</groupId>
	<artifactId>javaparser</artifactId>
	<version>${java-parser-version}</version>
	</dependency>
	</dependencies>
	<build>
	<plugins>
	<plugin>
	<artifactId>maven-compiler-plugin</artifactId>
	<configuration>
	<source>1.6</source>
	<target>1.6</target>
	</configuration>
	</plugin>
	</plugins>
	</build>
	</project>

view raw pom.xml hosted with ❤ by GitHub

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

	package org.fc;
	import japa.parser.JavaParser;
	import japa.parser.ast.CompilationUnit;
	import japa.parser.ast.expr.FieldAccessExpr;
	import japa.parser.ast.expr.MethodCallExpr;
	import japa.parser.ast.expr.NameExpr;
	import japa.parser.ast.stmt.BlockStmt;
	import japa.parser.ast.stmt.ExpressionStmt;
	import japa.parser.ast.stmt.Statement;
	import japa.parser.ast.visitor.VoidVisitorAdapter;
	import java.io.File;
	import java.io.FileInputStream;
	import java.io.FileOutputStream;
	import java.io.IOException;
	import java.io.InputStream;
	import java.io.PrintStream;
	import java.util.HashMap;
	import java.util.Iterator;
	import java.util.List;
	import java.util.Map;
	import org.apache.commons.io.FileUtils;
	import org.apache.commons.io.IOUtils;
	public class CleanupSysOut {
	File inputFile;
	File outputFile;
	Map<String, String> namespaceMap = new HashMap<String, String>();
	public CleanupSysOut(File file) {
	inputFile = file;
	}
	public CleanupSysOut(File in, File out) {
	inputFile = in;
	outputFile = out;
	}
	protected CompilationUnit getCompilationUnit() {
	CompilationUnit cu = null;
	InputStream in = null;
	try {
	in = new FileInputStream(inputFile);
	cu = JavaParser.parse(in);
	} catch (Exception e) {
	throw new RuntimeException(e);
	} finally {
	IOUtils.closeQuietly(in);
	}
	return cu;
	}
	protected void writeOutput(CompilationUnit cu) throws IOException {
	PrintStream out = new PrintStream(new FileOutputStream(outputFile));
	out.println(cu.toString());
	IOUtils.closeQuietly(out);
	}
	protected String debugPrint(CompilationUnit cu) throws IOException {
	return cu.toString();
	}
	// MethodVisitor
	private class MethodVisitor extends VoidVisitorAdapter&lt;Object&gt; {
	private boolean isSysout(MethodCallExpr methodCall) {
	if ( "println".equals(methodCall.getName()) ) {
	if ( methodCall.getScope() instanceof FieldAccessExpr ) {
	FieldAccessExpr fieldExpr = (FieldAccessExpr) methodCall.getScope();
	if ( "out".equals(fieldExpr.getField()) ) {
	if ( fieldExpr.getScope() instanceof NameExpr ) {
	NameExpr clazz = (NameExpr) fieldExpr.getScope();
	if ( "System".equals(clazz.getName()) ) {
	return true;
	}
	}
	}
	}
	}
	return false;
	}
	@Override
	public void visit(BlockStmt blockStmt, Object arg) {
	List<Statement> stmts = blockStmt.getStmts();
	if ( null != stmts ) {
	Iterator<Statement> itr = stmts.iterator();
	while ( itr.hasNext() ) {
	Statement stmt = itr.next();
	if ( stmt instanceof ExpressionStmt ) {
	if ( ((ExpressionStmt) stmt).getExpression() instanceof MethodCallExpr ) {
	MethodCallExpr methodCall = (MethodCallExpr) ((ExpressionStmt) stmt).getExpression();
	if ( isSysout(methodCall) ) {
	itr.remove();
	}
	}
	} else {
	stmt.accept(this, arg);
	}
	}
	}
	}
	}
	// MethodVisitor ends
	public void doCleanup() {
	try {
	CompilationUnit cu = getCompilationUnit();
	new MethodVisitor().visit(cu, null);
	if ( null != outputFile ) {
	writeOutput(cu);
	}
	} catch (Exception e) {
	throw new RuntimeException(e);
	}
	}
	static void cleanupDir(File inputDir, File outputDir) {
	for ( File java : FileUtils.listFiles(inputDir, new String[] { "java" }, true) ) {
	String relativePath = java.getAbsolutePath().replace(inputDir.getAbsolutePath(), "").substring(1);
	File outputFile = new File(outputDir, relativePath);
	outputFile.getParentFile().mkdirs();
	new CleanupSysOut(java, outputFile).doCleanup();
	}
	}
	public static void main(String[] args) {
	if ( args.length == 2 ) {
	cleanupDir(new File(args[0]), new File(args[1]));
	} else {
	System.err.println("Usage: java " + CleanupSysOut.class.getName() + " <input-dir> <output-dir>");
	}
	}
	}

view raw CleanupSysOut.java hosted with ❤ by GitHub

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 :)

	(defun luhn-sum (list n)
	(if (null list)
	0
	(+ (let ((x (car list)))
	(if (= 1 (mod n 2))
	(let ((y (* 2 x)))
	(if (> y 9)
	(+ 1 (mod y 10))
	y))
	x))
	(luhn-sum (cdr list) (+ 1 n)))
	)
	)
	(defun card-no-str-to-num (card-no)
	(mapcar (lambda (x) (string-to-number x 10)) (cdr (reverse (cdr (split-string card-no "")))))
	)
	(defun luhn-check (card-no)
	(eq 0 (mod (luhn-sum (card-no-str-to-num card-no) 0) 10))
	)
	(luhn-check "49927398716")

view raw luhn-check.el hosted with ❤ by GitHub