Sweep: fix comment whitespace issues in translate-utils.coffee

[nwinter]

In app/lib/translate-utils.coffee, some of the comments aren't indented correctly, instead just starting at the beginning of the line. We should fix the comment whitespace on all the improperly indented comments in this file.

[wwzeng1]

Hi @nwinter! Sweep timed out due to your repository size. We're increasing our timeouts, expect this to be fixed shortly!

[sweep-ai]

Here's the PR! #7057.

⚡ Sweep Free Trial: I used GPT-4 to create this ticket. You have 5 GPT-4 tickets left. For more GPT-4 tickets, visit our payment portal.

• Install Sweep Configs: Pull Request

---

Step 1: 🔍 Code Search

I found the following snippets in your repository. I will now analyze these snippets and come up with a plan.

Some code snippets I looked at (click to expand). If some file is missing from here, you can mention the path in the ticket description.

codecombat/app/lib/translate-utils.coffee

Lines 1 to 378 in 23d0134

	module.exports.translateJS = (jsCode, language='cpp', fullCode=true) ->
	return translateJSBrackets(jsCode, language, fullCode) if language in ['cpp', 'java']
	return translateJSWhitespace(jsCode, language) if language in ['python', 'lua', 'coffeescript']
	return jsCode if language is 'javascript'
	console.warn 'Unsupported language translation: from javascript to', language
	return jsCode
	translateJSBrackets = (jsCode, language='cpp', fullCode=true) ->
	# Supports cpp or java
	# Find all global statement(except global variables) and move into main function
	reorderGlobals = (strs) ->
	insertPlace = strs.length-1
	for i in [strs.length-2..0] by -1
	continue if /\n?function/.test(strs[i])
	continue if /^[ \t]+\/\//.test(strs[i])
	insertPlace -= 1
	strs.splice(insertPlace, 0, strs.splice(i,1)[0])
	mainLen = strs[strs.length-1].split('\n').length
	final = strs.splice(insertPlace).join('')
	finals = final.split('\n')
	insertPlace = finals.length - mainLen
	for i in [finals.length-mainLen-1..0] by -1
	continue unless finals[i]
	continue if /var /.test(finals[i])
	insertPlace -= 1
	finals.splice(insertPlace, 0, finals.splice(i,1)[0])
	return strs.concat([finals.slice(0, insertPlace).join('\n'), finals.slice(insertPlace).join('\n')])
	# Find header comments function definitions in order to hoist them out of the main function
	matchBrackets = (str, startIndex) ->
	cc = 0
	for i in [startIndex...str.length]
	cc += 1 if str[i] == '{'
	if str[i] == '}'
	cc -= 1
	return i+2 unless cc
	splitFunctions = (str) ->
	creg = /\n?[ \t]*[^\/]/
	startCommentReg = /^\n?(\/\/.*?\n)*\n/
	comments = startCommentReg.exec(str)
	if comments
	startComments = comments[0].slice(0, -1) # left the tailing \n
	str = str.slice startComments.length
	unless creg.exec str
	return [startComments, str]
	else
	strComments = ''
	indices = []
	reg = /\n(\/\/.*?\n)*function/gi
	indices.push 0 if str.startsWith("function ")
	while (result = reg.exec(str))
	indices.push result.index+1
	split = []
	end = 0
	# split.push {s: 0, e: indices[0]} if indices.length
	for i in indices
	split.push {s: end, e: i} if end != i
	end = matchBrackets str, i
	split.push {s: i, e: end}
	split.push {s: end, e: str.length}
	header = if startComments then [startComments] else []
	return header.concat(reorderGlobals(split.map (s) -> str.slice s.s, s.e ))
	jsCodes = splitFunctions jsCode
	if fullCode
	# Remove whitespace-only pieces, except for in the last piece
	jsCodes = _.filter(jsCodes.slice(0, jsCodes.length - 1), (piece) -> piece.replace(/\s/g, '').length).concat(jsCodes[jsCodes.length - 1])
	else
	# Remove all whitespace-only pieces
	jsCodes = _.filter jsCodes, (piece) -> piece.replace(/\s/g, '').length
	len = jsCodes.length
	if len
	lines = jsCodes[len-1].trimStart().split '\n'
	else
	lines = []
	#console.log "Split code segments into", _.cloneDeep(jsCodes)
	if fullCode
	if language is 'cpp'
	jsCodes[len-1] = """
	int main() {
	#{(lines.map (line) -> ' ' + line).join '\n'}
	return 0;
	}
	"""
	else if language is 'java'
	if len > 2 and !(/function/.test jsCodes[len-2])
	jsCodes[len-2] = (jsCodes[len-2].split('\n').map (line) ->
	if / = /.test line
	line = 'static ' + line
	line
	).join('\n')
	hasHeader = /^\/\//.test(jsCodes[0])
	startIndex = if hasHeader then 1 else 0
	functionLines = jsCodes.splice(startIndex, len - 1 - startIndex).join('\n').trimStart().split('\n')
	functionLines.shift() while functionLines.length and not functionLines[0] # Trim starting whitespace lines
	len = jsCodes.length
	jsCodes[len-1] = """
	public class AI {#{if functionLines.length then '\n' + (functionLines.map (line) -> ' ' + line).join '\n' else ''}
	public static void main(String[] args) {
	#{(lines.map (line) -> ' ' + line).join '\n'}
	}
	}
	"""
	else if len
	jsCodes[len-1] = (lines.map (line) -> ' ' + line).join('\n') # Add whitespace at beginning of each line to make regexes easier
	functionReturnType = if language is 'cpp' then 'auto' else 'public static var' # TODO: figure out some auto return types for Java
	functionParamType = if language is 'cpp' then 'auto' else 'Object' # TODO: figure out some auto/void param types for Java
	for i in [0...len]
	s = jsCodes[i]
	s = s.replace /function (.+?)\((.*?)\)/g, (match, functionName, functionParams) ->
	typedParameters = _.filter(functionParams.split(/, ?/)).map((e) -> "#{functionParamType} #{e}").join(', ')
	"#{functionReturnType} #{functionName}(#{typedParameters})"
	s = s.replace /var (x|y|z|dist)/g, 'float $1'
	s = s.replace /var (\w+)Index/g, 'int $1Index'
	s = s.replace /var (i|j|k)(?![a-zA-Z0-9_])/g, 'int $1'
	s = s.replace /\ ===\ /g, ' == '
	s = s.replace /\ !== /g, ' != '
	s = s.replace /hero\.throw\(/g, 'hero.throwEnemy('
	s = s.replace /\ = \[([^;]*)\];/g, ' = {$1};'
	if language is 'cpp'
	s = s.replace /\.length/g, '.size()'
	s = s.replace /\.push\(/g, '.push_back('
	s = s.replace /\.pop\(/g, '.pop_back('
	s = s.replace /\.shift\(/g, '.pop('
	s = s.replace /\ var /g, ' auto '
	s = s.replace /\(var /g, '(auto '
	s = s.replace /\nvar /g, '\nauto '
	s = s.replace /\ const /g, ' const auto '
	s = s.replace /\nconst /g, '\nconst auto '
	s = s.replace /\ return \[([^;]*)\];/g, ' return {$1};'
	# TODO: figure out how we are going to call other methods in Java
	# TODO: figure out how we are going to handle {x: 34, y: 30} object literals in Java
	# TODO: figure out how we are going to handle array methods in Java
	# Don't substitute these within comments
	noComment = '^( *[^/\\r\\n]*?)' # keep leading whitespace
	if language is 'cpp'
	newRegex = new RegExp(noComment + '([^*])new ', 'gm')
	while newRegex.test(s)
	s = s.replace newRegex, '$1$2*new '
	quotesReg = new RegExp(noComment + "'(.*?)'", 'gm')
	while quotesReg.test(s)
	s = s.replace quotesReg, '$1"$2"'
	# first replace ' to " then replace object
	if language is 'cpp'
	s = s.replace /\{\s*"?x"?\s*:\s*([^,]+),\s*"?y"?\s*:\s*([^\}]*)\}/g, '{$1, $2}' # {x:1, y:1} -> {1, 1}
	else if language is 'java'
	# Let's use Vectors instead of object literals in this case
	s = s.replace /\{\s*"?x"?\s*:\s*([^,]+),\s*"?y"?\s*:\s*([^\}]*)\}/g, 'new Vector($1, $2)' # {x:1, y:1} -> new Vector(1, 1)
	jsCodes[i] = s
	unless fullCode
	if len
	lines = jsCodes[len-1].split '\n'
	jsCodes[len-1] = (lines.map (line) -> line.slice 1).join('\n') # Remove leading convenience whitespace that we added
	else
	jsCodes = []
	jsCodes.join '\n'
	translateJSWhitespace = (jsCode, language='lua') ->
	# Supports python, lua, or coffeescript
	s = jsCode.split('\n').map((line) -> ' ' + line).join('\n') # Add whitespace at beginning of each line to make regexes easier
	if language is 'lua'
	s = s.replace /function (.+?)\((.*)\) ?{/g, 'function $1($2)' # Just remove the trailing {
	else if language is 'python'
	s = s.replace /function (.+?)\((.*)\) ?{/g, 'def $1($2):' # Convert trailing { to :
	else if language is 'coffeescript'
	s = s.replace /function (.+?)\((.*)\) ?{/g, (match, functionName, functionParams) ->
	if functionParams
	"#{functionName} = (#{functionParams}) ->"
	else
	"#{functionName} = ->"
	# Rewrite for-loops
	# for(i=0; i < archers.length; i++) {
	# var archer = archers[i];
	cStyleForInLoopWithVariableAssignmentRegex = /for ?\((?:var )?(.+?) ?= ?0; ?\1 ?< ?(.+?).length; ?(?:.*?\+\+.*?)\) *\{?\n(.*)var (.+?) ?= ?\2\[\1\];? *$/gm
	if language is 'lua'
	# for i, archer in pairs(archers) do
	s = s.replace cStyleForInLoopWithVariableAssignmentRegex, 'for $1, $4 in pairs($2) do'
	else if language is 'python'
	#s = s.replace cStyleForInLoopWithVariableAssignmentRegex, 'for $1, $4 in enumerate($2):' # I guess we usually do the other way for scaffolding learning similar to how we do it in JS instead of teaching enumerate
	s = s.replace cStyleForInLoopWithVariableAssignmentRegex, 'for $1 in range(len($2)):\n$3$4 = $2[$1]'
	else if language is 'coffeescript'
	# for archer in archers
	s = s.replace cStyleForInLoopWithVariableAssignmentRegex, 'for $4, $1 in $2'
	# for(i=0; i < archers.length; i++) {
	cStyleForInLoopRegex = /for ?\((?:var )?(.+?) ?= ?0; ?\1 ?< ?(.+?).length; ?(?:.*?\+\+.*?)\) *\{?/g
	if language is 'lua'
	# for i in pairs(archers) do
	s = s.replace cStyleForInLoopRegex, 'for $1 in pairs($2) do'
	else if language is 'python'
	# for i in range(0, len(archers)):
	s = s.replace cStyleForInLoopRegex, 'for $1 in range(len($2)):'
	else if language is 'coffeescript'
	# for i in [0...archers.length]
	s = s.replace cStyleForInLoopRegex, 'for $1 in [0...$2.length]'
	# for(i=0; i < 10; i++) {
	cStyleForLoopRegex = /for ?\((?:var )?(.+?) ?= ?(\d+); ?\1 ?< ?(.+?); ?(?:.*?\+\+.*?)\) *\{?/g
	if language is 'lua'
	# for i=0, 10 do
	s = s.replace cStyleForLoopRegex, 'for $1=$2, $3 do'
	else if language is 'python'
	# for i in range(0, 10):
	s = s.replace cStyleForLoopRegex, 'for $1 in range($2, $3):'
	else if language is 'coffeescript'
	# for i in [0...10]
	s = s.replace cStyleForLoopRegex, 'for $1 [$2...$3]'
	# for(y=110; y >= 38; i -= 18) {
	# This is brittle and will not get inclusive vs. exclusive ranges right, but better than nothing
	cStyleForLoopWithArithmeticRegex = /for ?\((?:var )?(.+?) ?= ?(\d+); ?\1 ?(<=|<|>=|>) ?(.+?); ?\1 ?\+?(-?)= ?(.*)\) *\{?/g
	if language is 'lua'
	# for y=110, 38, -18 do
	s = s.replace cStyleForLoopWithArithmeticRegex, 'for $1=$2, $4, $5$6 do'
	else if language is 'python'
	# for y in range(110, 38, -18):
	s = s.replace cStyleForLoopWithArithmeticRegex, 'for $1 in range($2, $4, $5$6):'
	else if language is 'coffeescript'
	# for y in [110...38, -18]
	s = s.replace cStyleForLoopWithArithmeticRegex, 'for $1 [$2...$4, $5$6]'
	# There are a lot of other for-loop possibilities, but we'll handle those with manual solutions
	if language is 'lua'
	s = s.replace /\ ===\ /g, ' == '
	s = s.replace /\ !==? /g, ' ~= '
	s = s.replace /(\S+)(\+|-){2}/g, '$1 = $1 $2 1' # Rewrite postfix ++ and --, like count++ -> count = count + 1
	s = s.replace /(\+|-){2}(\S+)/g, '$2 = $2 $1 1' # Rewrite prefix ++ and --, like ++count -> count = count + 1
	s = s.replace /(\S+) ?(\+|-|\*|\/)= ?(.+)/g, '$1 = $1 $2 $3' # Rewrite +=, -=, etc.
	else if language is 'coffeescript'
	s = s.replace /\ ===?\ /g, ' is '
	s = s.replace /\ !==? /g, ' isnt '
	else if language is 'python'
	s = s.replace /\ ===?\ /g, ' == ' # Maybe we should rewrite to `is` instead?
	s = s.replace /\ !==? /g, ' != '
	s = s.replace /(\S+)(\+|-){2}/g, '$1 $2= 1' # Rewrite postfix ++ and --, like count++ -> count += 1
	s = s.replace /(\+|-){2}(\S+)/g, '$2 $1= 1' # Rewrite prefix ++ and --, like ++count -> count += 1
	s = s.replace /\ &&\ /g, ' and '
	s = s.replace /\ \|\|\ /g, ' or '
	s = s.replace /\!([$A-Z_(])/gi, 'not $1'
	if language is 'python'
	s = s.replace /\.push\(/g, '.append('
	s = s.replace /\.shift\(0?\)/g, '.pop(0)'
	else if language is 'lua'
	s = s.replace /\.push\(/g, '.insert('
	s = s.replace /\.pop\(/g, '.remove('
	s = s.replace /\.shift\(0?\)/g, '.remove(0)'
	if language is 'lua'
	s = s.replace /\ var /g, ' local '
	s = s.replace /\ = \[([^;]*)\];/g, ' = {$1};'
	s = s.replace /\(var /g, '(local '
	s = s.replace /\nvar /g, '\nlocal '
	s = s.replace /\ return \[([^;]*)\];/g, ' return {$1};'
	else if language in ['python', 'coffeescript']
	s = s.replace /^ *var [^=\n]*$\n/gm, '' # Remove variable declarations without initialization
	s = s.replace /\ var /g, ' '
	s = s.replace /\(var /g, '('
	s = s.replace /\nvar /g, '\n'
	# Don't substitute these within comments
	noComment = '^ *([^/\\r\\n]*?)'
	if language in ['python', 'lua']
	newRegex = new RegExp(noComment + 'new ', 'gm')
	while newRegex.test(s)
	s = s.replace newRegex, '$1'
	# Rewrite comments
	commentStart = commentStarts[language] or '#'
	commentStartRegex = new RegExp "([ \t]*?)//", 'gm'
	s = s.replace commentStartRegex, "$1#{commentStart}" # ` // Comment` -> ` # Comment`
	# No semicolons
	s = s.replace /;/g, ''
	# For Lua, replace periods with colons for method calls (but not other property accesses)
	if language is 'lua'
	s = s.replace /([$A-Z_][$0-9a-z_]*)\.([$A-Za-z_][0-9A-Za-z_$]*)\(/gi, '$1:$2('
	# We still use periods for Math (still using the JavaScript library), it's static vs. instance method thing
	# Hack: re-replace back to dots in those cases by looking at initial capital letter of variable name
	s = s.replace /([$A-Z_][$0-9a-z_]*):([$A-Za-z_][0-9A-Za-z_$]*)\(/g, '$1.$2('
	# Rewrite while loops
	if language is 'lua'
	s = s.replace /^(\s*while) ?\((.*)\) ?\{?/gm, '$1 $2 do'
	else if language is 'python'
	s = s.replace /^(\s*while) ?\((.*)\) ?\{?/gm, '$1 $2:'
	else if language is 'coffeescript'
	s = s.replace /^(\s*while) ?\((.*)\) ?\{?/gm, '$1 $2'
	# Rewrite if conditions
	if language is 'lua'
	s = s.replace /else if/g, 'elseif'
	s = s.replace /(} *)?( *(if|elseif)) ?\((.*)\) ?\{?/gm, '$2 $4 then'
	else if language is 'python'
	s = s.replace /else if/g, 'elif'
	s = s.replace /(} *)?( *(if|elif)) ?\((.*)\) ?\{?/gm, '$2 $4:'
	s = s.replace /(}\s*)?else\s*{/g, 'else:'
	else if language is 'coffeescript'
	s = s.replace /(} *)?( *(if|else if)) ?\((.*)\) ?\{?/gm, '$2 $4'
	s = s.replace /(}\s*)?else\s*{/g, 'else'
	# Rewrite else { to else
	s = s.replace /(}\s*)?else\s*{/g, 'else'
	if language is 'lua'
	# Rewrite standalone `}` to `end`
	s = s.replace /^(\s*)\} *$/gm, '$1end'
	# Remove `end` as part of part of if/elseif/else chains
	s = s.replace /^(\s*)end ?}?\n((\n|\s|--.*\n)*^\1)(elseif|else)/gm, '$2$4' # The ^\1 only matches the same level of indentation
	else if language in ['python', 'coffeescript']
	# Remove stanadlone `}`
	s = s.replace /\n\s*\} *$/gm, ''
	if language is 'lua'
	s = s.replace /null/g, 'nil'
	s = s.replace /(\S+)\.length/g, '#$1' # Do this after if/else paren/bracket replacement
	else if language is 'python'
	s = s.replace /true/g, 'True'
	s = s.replace /false/g, 'False'
	s = s.replace /null/g, 'None'
	s = s.replace /(\S+)\.length/g, 'len($1)' # Do this after if/else paren/bracket replacement
	if language is 'coffeescript'
	# Remove unnecessary parenthesis in CofeeScript
	s = s.replace /([$A-Z_][0-9A-Z_$]*)\(([^()]+)\)(?!\))$/gim, '$1 $2'
	# Use simple loops in CoffeeScript
	s = s.replace /while true$/gm, 'loop'
	if language is 'lua'
	# Convert : to =. {x:1, y:1} -> {x=1, y=1}
	s = s.replace /\{\s*['"]?(\S+?)['"]?\s*:\s*([^,]+)\}/g, '{$1=$2}' # 1 element
	s = s.replace /\{\s*['"]?(\S+?)['"]?\s*:\s*([^,]+),\s*['"]?(\S+?)['"]?\s*:\s*([^\}]*)\}/g, '{$1=$2, $3=$4}' # 2 elements
	s = s.replace /\{\s*['"]?(\S+?)['"]?\s*:\s*([^,]+),\s*['"]?(\S+?)['"]?\s*:\s*([^\}]*),\s*['"]?(\S+?)['"]?\s*:\s*([^\}]*)\}/g, '{$1=$2, $3=$4, $4=6}' # 3 elements
	# TODO: something flexible for arbitrary n elements
	else if language is 'python'
	# Add quotes. {x:1, y:1} -> {"x": 1, "y": 1}
	s = s.replace /\{\s*['"]?(\S+?)['"]?\s*:\s*([^,]+)\}/g, '{"$1": $2}' # 1 element
	s = s.replace /\{\s*['"]?(\S+?)['"]?\s*:\s*([^,]+),\s*['"]?(\S+?)['"]?\s*:\s*([^\}]*)\}/g, '{"$1": $2, "$3": $4}' # 2 elements
	s = s.replace /\{\s*['"]?(\S+?)['"]?\s*:\s*([^,]+),\s*['"]?(\S+?)['"]?\s*:\s*([^\}]*),\s*['"]?(\S+?)['"]?\s*:\s*([^\}]*)\}/g, '{"$1": $2, "$3": $4, "$5": $6}' # 3 elements
	# TODO: something flexible for arbitrary n elements
	if language is 'lua'
	# Try incrementing all literal array indexes under, say, 10 by 1 to offset 1-based indexing. Hack, but most of those levels will need manual attention anyway.
	s = s.replace /\[(\d)\]/g, (match, index) -> "[#{parseInt(index, 10) + 1}]"
	# TODO: see if we can do something about lack of a continue statement in Lua? Maybe too hard and we should give up.
	lines = s.split '\n'
	output = (lines.map (line) -> line.slice 1).join('\n') # Remove leading convenience whitespace that we added
	output
	# Note: These need to be double-escaped for insertion into regexes
	commentStarts =
	javascript: '//'
	python: '#'
	coffeescript: '#'
	lua: '--'
	java: '//'
	cpp: '//'
	html: '<!--'
	css: '/\\*'

codecombat/app/views/play/level/tome/Spell.coffee

Lines 53 to 176 in 23d0134

	@setLanguage codeLanguage
	else if @otherSession and @team is @otherSession.get 'team'
	@setLanguage @otherSession.get('submittedCodeLanguage') or @otherSession.get('codeLanguage')
	else
	@setLanguage 'javascript'
	@source = @originalSource
	@parameters = p.parameters
	if @otherSession and @team is @otherSession.get('team') and sessionSource = @otherSession.getSourceFor(@spellKey)
	# Load opponent code from other session (new way, not relying on PlayLevelView loadOpponentTeam)
	@source = replaceSimpleLoops sessionSource, @language
	else if @permissions.readwrite.length and sessionSource = @session.getSourceFor(@spellKey)
	# Load either our code or opponent code (old way, opponent code copied into our session in PlayLevelView loadOpponentTeam)
	if sessionSource isnt '// Should fill in some default source\n' # TODO: figure out why session is getting this default source in there and stop it
	@source = replaceSimpleLoops sessionSource, @language
	if p.aiSource and not @otherSession and not @canWrite()
	@source = @originalSource = p.aiSource
	@isAISource = true
	destroy: ->
	@view?.destroy()
	@topBarView?.destroy()
	@thang = null
	@worker = null
	setLanguage: (@language) ->
	@language = 'html' if @level.isType('web-dev')
	@displayCodeLanguage = utils.capitalLanguages[@language]
	if @language in ['cpp', 'java', 'lua', 'coffeescript', 'python'] and not @languages[@language]
	@languages[@language] = translateJS @languages.javascript, @language
	@originalSource = @languages[@language] ? @languages.javascript
	@originalSource = @addPicoCTFProblem() if window.serverConfig.picoCTF
	if @level.isType('web-dev')
	# Pull apart the structural wrapper code and the player code, remember the wrapper code, and strip indentation on player code.
	playerCode = utils.extractPlayerCodeTag(@originalSource)
	@wrapperCode = @originalSource.replace /<playercode>[\s\S]*<\/playercode>/, '☃' # ☃ serves as placeholder for constructHTML
	@originalSource = playerCode
	# Translate comments chosen spoken language.
	return unless @commentContext
	context = $.extend true, {}, @commentContext
	spokenLanguage = me.get 'preferredLanguage'
	@originalSource = @translateCommentContext source: @originalSource, commentContext: @commentContext, commentI18N: @commentI18N, spokenLanguage: spokenLanguage, codeLanguage: @language
	@wrapperCode = @translateCommentContext source: @wrapperCode, commentContext: @commentContext, commentI18N: @commentI18N, spokenLanguage: spokenLanguage, codeLanguage: @language
	if /loop/.test(@originalSource) and @level.isType('course', 'course-ladder', 'hero', 'hero-ladder')
	# Temporary hackery to make it look like we meant while True: in our sample code until we can update everything
	@originalSource = replaceSimpleLoops @originalSource, @language
	translateCommentContext: ({ source, commentContext, commentI18N, codeLanguage, spokenLanguage }) ->
	commentContext = $.extend true, {}, commentContext
	if codeLanguage is 'lua'
	for k, v of commentContext
	commentContext[k] = v.replace /\b([a-zA-Z]+)\.([a-zA-Z_]+\()/, '$1:$2'
	if commentI18N
	while spokenLanguage
	spokenLanguage = spokenLanguage.substr 0, spokenLanguage.lastIndexOf('-') if fallingBack?
	if spokenLanguageContext = commentI18N[spokenLanguage]?.context
	commentContext = _.merge commentContext, spokenLanguageContext
	break
	fallingBack = true
	try
	translatedSource = _.template source, commentContext
	catch e
	console.error "Couldn't create example code template of", source, "\nwith commentContext", commentContext, "\nError:", e
	translatedSource = source
	translatedSource
	untranslateCommentContext: ({ source, commentContext, commentI18N, codeLanguage, spokenLanguage }) ->
	commentContext = $.extend true, {}, commentContext
	if codeLanguage is 'lua'
	for k, v of commentContext
	commentContext[k] = v.replace /\b([a-zA-Z]+)\.([a-zA-Z_]+\()/, '$1:$2'
	if commentI18N
	while spokenLanguage
	spokenLanguage = spokenLanguage.substr 0, spokenLanguage.lastIndexOf('-') if fallingBack?
	if spokenLanguageContext = commentI18N[spokenLanguage]?.context
	commentContext = _.merge commentContext, spokenLanguageContext
	break
	fallingBack = true
	for k, v of commentContext
	source = source.replace v, "<%= #{k} %>"
	source
	getSolution: (codeLanguage) ->
	hero = _.find (@level.get('thangs') ? []), id: 'Hero Placeholder'
	component = _.find(hero.components ? [], (x) -> x?.config?.programmableMethods?.plan)
	plan = component.config?.programmableMethods?.plan
	solutions = _.filter (plan?.solutions ? []), (s) -> not s.testOnly and s.succeeds
	rawSource = _.find(solutions, language: codeLanguage)?.source
	rawSource
	constructHTML: (source) ->
	@wrapperCode.replace '☃', source
	addPicoCTFProblem: ->
	return @originalSource unless problem = @level.picoCTFProblem
	description = """
	-- #{problem.name} --
	#{problem.description}
	""".replace /<p>(.*?)<\/p>/gi, '$1'
	("// #{line}" for line in description.split('\n')).join('\n') + '\n' + @originalSource
	addThang: (thang) ->
	if @thang?.thang.id is thang.id
	@thang.thang = thang
	else
	@thang = {thang: thang, aether: @createAether(thang), castAether: null}
	removeThangID: (thangID) ->
	@thang = null if @thang?.thang.id is thangID
	canRead: (team) ->
	(team ? me.team) in @permissions.read or (team ? me.team) in @permissions.readwrite
	canWrite: (team) ->
	(team ? me.team) in @permissions.readwrite
	getSource: ->

codecombat/app/collections/Levels.coffee

Lines 1 to 58 in 23d0134

	CocoCollection = require 'collections/CocoCollection'
	Level = require 'models/Level'
	utils = require 'core/utils'
	translateUtils = require 'lib/translate-utils'
	module.exports = class LevelCollection extends CocoCollection
	url: '/db/level'
	model: Level
	fetchForClassroom: (classroomID, options={}) ->
	options.url = "/db/classroom/#{classroomID}/levels"
	@fetch(options)
	fetchForClassroomAndCourse: (classroomID, courseID, options={}) ->
	options.url = "/db/classroom/#{classroomID}/courses/#{courseID}/levels"
	@fetch(options)
	fetchForCampaign: (campaignSlug, options={}) ->
	options.url = "/db/campaign/#{campaignSlug}/levels"
	@fetch(options)
	getSolutionsMap: (languages) ->
	@models.reduce((map, level) =>
	targetLangs = if level.get('primerLanguage') then [level.get('primerLanguage')] else languages
	allSolutions = _.filter level.getSolutions(), (s) -> not s.testOnly
	solutions = @constructor.getSolutionsHelper({ targetLangs, allSolutions })
	map[level.get('original')] = solutions?.map((s) => {source: @fingerprint(s.source, s.language), description: s.description})
	map
	, {})
	@getSolutionsHelper: ({ targetLangs, allSolutions }) ->
	solutions = []
	for lang in targetLangs
	if lang is 'html'
	for s in allSolutions when s.language is 'html'
	strippedSource = utils.extractPlayerCodeTag(s.source or '')
	s.source = strippedSource if strippedSource
	solutions.push s
	else if lang isnt 'javascript' and not _.find(allSolutions, language: lang)
	for s in allSolutions when s.language is 'javascript'
	s.language = lang
	s.source = translateUtils.translateJS(s.source, lang)
	solutions.push s
	else
	for s in allSolutions when s.language is lang
	solutions.push s
	solutions
	fingerprint: (code, language) ->
	# Add a zero-width-space at the end of every comment line
	@constructor.fingerprintHelper(code, language)
	@fingerprintHelper: (code, language) ->
	switch language
	when ['javascript', 'java', 'cpp'] then code.replace /^(\/\/.*)/gm, "$1​"
	when 'lua' then code.replace /^(--.*)/gm, "$1​"
	when 'html' then code.replace /^(<!--.*)-->/gm, "$1​-->"
	else code.replace /^(#.*)/gm, "$1​"

codecombat/app/models/LevelSession.coffee

Lines 97 to 171 in 23d0134

	thresholdAchieved = Level.thresholdForScore(_.assign(_.pick(sessionScore, 'score', 'type'), {level}))
	if thresholdAchieved
	sessionScore.thresholdAchieved = thresholdAchieved
	scores
	@fakeID: 'ateacherfakesessionidval' # 24 characters like other IDs for schema validation
	isFake: -> @id is LevelSession.fakeID
	inLeague: (leagueId) ->
	return false unless @get('leagues')
	for league in @get('leagues')
	return true if league.leagueID is leagueId
	return false
	updateKeyValueDb: (keyValueDb) ->
	oldDb = @get('keyValueDb') ? {}
	@originalKeyValueDb ?= oldDb
	@set('keyValueDb', keyValueDb) if _.size keyValueDb
	saveKeyValueDb: ->
	keyValueDb = @get('keyValueDb') ? {}
	return unless @originalKeyValueDb
	return if @isFake()
	for key, value of keyValueDb
	oldValue = @originalKeyValueDb[key]
	continue if value is oldValue
	if not oldValue? or typeof(oldValue) is 'string' or typeof(value) is 'string'
	api.levelSessions.setKeyValue({ sessionID: @id, key, value})
	else if typeof(oldValue) is 'number' and typeof(value) is 'number'
	increment = value - oldValue
	api.levelSessions.incrementKeyValue({ sessionID: @id, key, value: increment})
	@set('keyValueDb', keyValueDb) if _.size keyValueDb
	delete @originalKeyValueDb
	countOriginalLinesOfCode: (level) ->
	# Count non-whitespace, non-comment lines starting at first unique code line
	# TODO: diff better to find truly changed lines
	sampleCodeByLanguage = level.getSampleCode @get('team')
	sampleCode = sampleCodeByLanguage[@get('codeLanguage')] ? sampleCodeByLanguage.html ? sampleCodeByLanguage.python ? ''
	sampleCode = sampleCode.replace @singleLineCommentOnlyRegex(), ''
	sampleCodeLines = sampleCode.split(/\n+/)
	sampleCodeLines = _.filter sampleCodeLines
	thang = if @get('team') is 'ogres' then 'hero-placeholder-1' else 'hero-placeholder'
	code = @getSourceFor("#{thang}/plan") ? ''
	code = code.replace @singleLineCommentOnlyRegex(), ''
	codeLines = code.split(/\n+/)
	codeLines = _.filter codeLines
	i = 0
	while i < codeLines.length and i < sampleCodeLines.length
	break if codeLines[i] isnt sampleCodeLines[i]
	++i
	count = codeLines.length - i
	#console.log "Got", count, "original lines from\n", code, "\n-----------\n", sampleCode
	Math.min count, 1000
	singleLineCommentOnlyRegex: ->
	if @get('codeLanguage') is 'html'
	commentStart = "#{commentStarts.html}|#{commentStarts.css}|#{commentStarts.javascript}"
	else
	commentStart = commentStarts[@get('codeLanguage')] or '#'
	new RegExp "^[ \t]*(#{commentStart}).*$", 'gm'
	# Note: These need to be double-escaped for insertion into regexes
	commentStarts =
	javascript: '//'
	python: '#'
	coffeescript: '#'
	lua: '--'
	java: '//'
	cpp: '//'
	html: '<!--'
	css: '/\\*'

codecombat/test/app/lib/translate_utils.spec.ozar.coffee

Lines 1 to 80 in 23d0134

	solutionsByLanguage = javascript: {}, lua: {}, python: {}, coffeescript: {}, java: {}, cpp: {}
	solutionsByLanguage.javascript.dungeonsOfKithgard = """
	// Move towards the gem.
	// Don’t touch the spikes!
	// Type your code below and click Run when you’re done.
	hero.moveRight();
	hero.moveDown();
	hero.moveRight();
	"""
	solutionsByLanguage.javascript.peekABoom = """
	// Build traps on the path when the hero sees a munchkin!
	while(true) {
	var enemy = hero.findNearestEnemy();
	if(enemy) {
	// Build a "fire-trap" at the Red X (41, 24)
	hero.buildXY("fire-trap", 41, 24);
	}
	// Add an else below to move back to the clearing
	else {
	// Move to the Wooden X (19, 19)
	hero.moveXY(19, 19);
	}
	}
	"""
	solutionsByLanguage.javascript.woodlandCleaver = """
	// Use your new "cleave" skill as often as you can.
	hero.moveXY(23, 23);
	while(true) {
	var enemy = hero.findNearestEnemy();
	if(hero.isReady("cleave")) {
	// Cleave the enemy!
	hero.cleave(enemy);
	} else {
	// Else (if cleave isn't ready), do your normal attack.
	hero.attack(enemy);
	}
	}
	"""
	solutionsByLanguage.javascript.aFineMint = """
	// Peons are trying to steal your coins!
	// Write a function to squash them before they can take your coins.
	function pickUpCoin() {
	var coin = hero.findNearestItem();
	if(coin) {
	hero.moveXY(coin.pos.x, coin.pos.y);
	}
	}
	// Write the attackEnemy function below.
	// Find the nearest enemy and attack them if they exist!
	function attackEnemy() {
	var enemy = hero.findNearestEnemy();
	if(enemy) {
	hero.attack(enemy);
	}
	}
	while(true) {
	attackEnemy(); // Δ Uncomment this line after you write an attackEnemy function.
	pickUpCoin();
	}
	"""
	solutionsByLanguage.javascript.libraryTactician = """
	// Hushbaum has been ambushed by ogres!
	// She is busy healing her soldiers, you should command them to fight!
	// The ogres will send more troops if they think they can get to Hushbaum or your archers, so keep them inside the circle!
	var archerTarget = null;
	// Soldiers spread out in a circle and defend.
	function commandSoldier(soldier, soldierIndex, numSoldiers) {
	var angle = Math.PI * 2 * soldierIndex / numSoldiers;
	var defendPos = {x: 41, y: 40};
	defendPos.x += 10 * Math.cos(angle);
	defendPos.y += 10 * Math.sin(angle);
	hero.command(soldier, "defend", defendPos);
	}

---

Step 2: 🧐 Snippet Analysis

From looking through the relevant snippets, I decided to make the following modifications:

File Path	Proposed Changes
app/lib/translate-utils.coffee	Go through the file and adjust the indentation for each comment that is not properly indented. Ensure that each comment starts at the same column as the code it is commenting on.

---

Step 3: 📝 Planning

I have created a plan for writing the pull request. I am now working my plan and coding the required changes to address this issue. Here is the planned pull request:

Fix comment whitespace issues in translate-utils.coffee
sweep/fix-comment-whitespace

Description

This PR fixes the comment whitespace issues in the translate-utils.coffee file. Some of the comments were not indented correctly and were starting at the beginning of the line. This PR adjusts the indentation of the comments to align with the code they are associated with.

Changes Made

• Adjusted the indentation of comments in the translate-utils.coffee file to align with the code they are commenting on.

Testing

No functional changes were made, so no additional testing is required.

Related Issue

Fixes #[issue_number]

---

Step 4: ⌨️ Coding

I have finished coding the issue. I am now reviewing it for completeness.

---

Step 5: 🔁 Code Review

Success! 🚀

---

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!
^{Join Our Discord}

[wwzeng1]

sweep: retry