Add support for Snakemake

boyter · May 6, 2024 · ca6a31b · ca6a31b
1 parent 8b85895
commit ca6a31b
Show file tree

Hide file tree

Showing 6 changed files with 154 additions and 17 deletions.
diff --git a/LANGUAGES.md b/LANGUAGES.md
@@ -151,7 +151,7 @@ License (license,licence,copying,copying3,unlicense,unlicence,license-mit,licenc
 Lisp (lisp,lsp)
 LLVM IR (ll)
 LOLCODE (lol,lols)
-Lua (lua)
+Lua (lua,luau)
 Luau (luau)
 Lucius (lucius)
 Luna (luna)
@@ -205,7 +205,7 @@ PRQL (prql)
 PSL Assertion (psl)
 Puppet (pp)
 PureScript (purs)
-Python (py)
+Python (py,pyw,pyi)
 Q# (qs)
 QCL (qcl)
 QML (qml)
@@ -232,6 +232,7 @@ Shell (sh,.tcshrc)
 Sieve (sieve)
 SKILL (il)
 Smarty Template (tpl)
+Snakemake (smk,rules,snakefile)
 SNOBOL (sno)
 Softbridge Basic (sbl)
 Solidity (sol)

diff --git a/SCC-OUTPUT-REPORT.html b/SCC-OUTPUT-REPORT.html
@@ -18,7 +18,7 @@
 		<th>453</th>
 		<th>7424</th>
 		<th>1516</th>
-		<th>396673</th>
+		<th>398173</th>
 		<th>3929</th>
 	</tr><tr>
 		<td>processor/workers_test.go</td>
@@ -250,16 +250,6 @@
 		<td>0</td>
 	    <td>2222</td>
 		<td>35</td>
-	</tr><tr>
-		<td>processor/cocomo_test.go</td>
-		<td></td>
-		<td>37</td>
-		<td>8</td>
-		<td>4</td>
-		<td>25</td>
-		<td>6</td>
-	    <td>699</td>
-		<td>23</td>
 	</tr><tr>
 		<td>processor/bloom.go</td>
 		<td></td>
@@ -270,6 +260,16 @@
 		<td>2</td>
 	    <td>1051</td>
 		<td>29</td>
+	</tr><tr>
+		<td>processor/cocomo_test.go</td>
+		<td></td>
+		<td>37</td>
+		<td>8</td>
+		<td>4</td>
+		<td>25</td>
+		<td>6</td>
+	    <td>699</td>
+		<td>23</td>
 	</tr><tr>
 		<td>processor/helpers.go</td>
 		<td></td>
@@ -318,7 +318,7 @@
 		<td>0</td>
 		<td>4</td>
 		<td>0</td>
-	    <td>170185</td>
+	    <td>171685</td>
 		<td>5</td>
 	</tr></tbody>
 	<tfoot><tr>
@@ -329,7 +329,7 @@
 		<th>453</th>
 		<th>7424</th>
 		<th>1516</th>
-    	<th>396673</th>
+    	<th>398173</th>
 		<th>3929</th>
 	</tr>
 	<tr>

diff --git a/examples/language/snakefile.smk b/examples/language/snakefile.smk
@@ -0,0 +1,67 @@
+# 67 lines 50 code 4 comments 13 blanks
+"""
+A sample Snakefile for testing line counting
+"""
+
+SAMPLES = ["A", "B"]
+
+
+# This is a
+# multiline
+# comment
+rule all:
+    input:
+        "plots/quals.svg"
+
+
+'''Sometimes even some
+comments in single quote
+fences.'''
+rule bwa_map:
+    input:
+        "data/genome.fa",  # Inline comments are also supported
+        "data/samples/{sample}.fastq"
+    output:
+        "mapped_reads/{sample}.bam"
+    shell:
+        "bwa mem {input} | samtools view -Sb - > {output}"
+
+
+rule samtools_sort:
+    input:
+        "mapped_reads/{sample}.bam"
+    output:
+        "sorted_reads/{sample}.bam"
+    shell:
+        "samtools sort -T sorted_reads/{wildcards.sample} "
+        "-O bam {input} > {output}"
+
+
+rule samtools_index:
+    input:
+        "sorted_reads/{sample}.bam"
+    output:
+        "sorted_reads/{sample}.bam.bai"
+    shell:
+        "samtools index {input}"
+
+
+rule bcftools_call:
+    input:
+        fa="data/genome.fa",
+        bam=expand("sorted_reads/{sample}.bam", sample=SAMPLES),
+        bai=expand("sorted_reads/{sample}.bam.bai", sample=SAMPLES)
+    output:
+        "calls/all.vcf"
+    shell:
+        "bcftools mpileup -f {input.fa} {input.bam} | "
+        "bcftools call -mv - > {output}"
+
+
+rule plot_quals:
+    input:
+        "calls/all.vcf"
+    output:
+        "plots/quals.svg"
+    script:
+        "scripts/plot-quals.py"
diff --git a/languages.json b/languages.json
@@ -5979,6 +5979,75 @@
       "python3"
     ]
   },
+  "Snakemake": {
+    "complexitychecks": [
+      "for ",
+      "for(",
+      "while ",
+      "while(",
+      "if ",
+      "if(",
+      "elif ",
+      "elif(",
+      "else ",
+      "else:",
+      "match ",
+      "match(",
+      "try ",
+      "try:",
+      "except ",
+      "except(",
+      "finally ",
+      "finally:",
+      "with ",
+      "with (",
+      "and ",
+      "and(",
+      "or ",
+      "or("
+    ],
+    "extensions": [
+      "smk",
+      "rules"
+    ],
+    "line_comment": [
+      "#"
+    ],
+    "multi_line": [],
+    "quotes": [
+      {
+        "end": "\"",
+        "start": "\""
+      },
+      {
+        "end": "'",
+        "start": "'"
+      },
+      {
+        "docString": true,
+        "end": "\"\"\"",
+        "start": "\"\"\""
+      },
+      {
+        "docString": true,
+        "end": "'''",
+        "start": "'''"
+      },
+      {
+        "docString": true,
+        "end": "\"\"\"",
+        "start": "r\"\"\""
+      },
+      {
+        "docString": true,
+        "end": "'''",
+        "start": "r'''"
+      }
+    ],
+    "filenames": [
+      "snakefile"
+    ]
+  },
   "PRQL": {
     "complexitychecks": [
       "case ",

diff --git a/processor/constants.go b/processor/constants.go
diff --git a/test-all.sh b/test-all.sh
@@ -883,7 +883,7 @@ else
 fi
 
 # Try out specific languages
-for i in 'Bosque ' 'Flow9 ' 'Bitbucket Pipeline ' 'Docker ignore ' 'Q# ' 'Futhark ' 'Alloy ' 'Wren ' 'Monkey C ' 'Alchemist ' 'Luna ' 'ignore ' 'XML Schema ' 'Web Services' 'Go ' 'Java ' 'Boo ' 'License ' 'BASH ' 'C Shell ' 'Korn Shell ' 'Makefile ' 'Shell ' 'Zsh ' 'Rakefile ' 'Gemfile ' 'Dockerfile ' 'Yarn ' 'Sieve ' 'F# ' 'Elm ' 'Terraform ' 'Clojure ' 'C# ' 'LLVM IR ' 'HAML ' 'FXML ' 'DM ' 'Nushell ' 'Racket ' 'DOT ' 'YAML ' 'Teal ' 'FSL ' 'INI ' 'Hare ' 'Templ ' 'Cuda ' 'GraphQL ' 'Bicep ' 'Pkl ' 'TypeSpec ' 'LALRPOP '
+for i in 'Bosque ' 'Flow9 ' 'Bitbucket Pipeline ' 'Docker ignore ' 'Q# ' 'Futhark ' 'Alloy ' 'Wren ' 'Monkey C ' 'Alchemist ' 'Luna ' 'ignore ' 'XML Schema ' 'Web Services' 'Go ' 'Java ' 'Boo ' 'License ' 'BASH ' 'C Shell ' 'Korn Shell ' 'Makefile ' 'Shell ' 'Zsh ' 'Rakefile ' 'Gemfile ' 'Dockerfile ' 'Yarn ' 'Sieve ' 'F# ' 'Elm ' 'Terraform ' 'Clojure ' 'C# ' 'LLVM IR ' 'HAML ' 'FXML ' 'DM ' 'Nushell ' 'Racket ' 'DOT ' 'YAML ' 'Teal ' 'FSL ' 'INI ' 'Hare ' 'Templ ' 'Cuda ' 'GraphQL ' 'Bicep ' 'Pkl ' 'TypeSpec ' 'LALRPOP ' 'Snakemake '
 do
     if ./scc "examples/language/" | grep -q "$i "; then
         echo -e "${GREEN}PASSED $i Language Check"