Merge pull request #80 from worldbank/v1.02

v1.02

admin/checklist-submitting-SSC.md

@@ -9,7 +9,7 @@
 	- [ ] 3.2 **Update version and date** - In the _version_ branch, update the version number and date in all ado-files and all dates in all help files. See section below for details.
 	- [ ] 3.3 **Update version globals** - Update the _version_ado_ local in the file _lint.ado_ and the  _VERSION_ global in _stata_linter_detect.py_ and _stata_linter_correct.py_.
 	- [ ] 3.4 **Update version in .pkg and .toc** - This has nothing to do with SSC but should be kept up to date to. This is for when people install directly through GitHub using `net install`. If any new command has been added, remember to add the files for that command to the `.pkg` file.
-	- [ ] 3.5 **Create a .zip file** - Create a .zip file with all ado-files and help files only. These files are not allowed to be in a sub-folder in this .zip file. No other files should be in this folder. Make a copy of this file in the archive folder of this package.
+	- [ ] 3.5 **Create a .zip file** - Create a .zip file with the files listed below (ado-files, Python scripts, and help files). If a version update ever includes a new ado-file or Python script necessary to run the linter, include that new file in the .zip too. These files are not allowed to be in a sub-folder in this .zip file. No other files should be in this folder. Make a copy of this file in the archive folder of this package.
 - [ ] 4. **Email Prof. Baum** - Email the .zip file created in step 3.5 to **[email protected]**.
 	- [ ] 4.1 - If any commands are added or deleted, make note of that in the email.
 	- [ ] 4.2 - If any of the meta info (title, description, keywords, version or author/contact) has changed then include those updates in your email.

admin/ssc-meta-info.md

@@ -11,7 +11,7 @@ The package contains a command that detects bad Stata coding practices in a do-f
 The command can also correct some of the issues flagged in a new do-file.
 The purpose of the command is to help users improve code clarity, readability, and organization in Stata do-files.
 This linter is based on the best practices outlined in The DIME Analytics Coding Guide published as an appendix to the book Development Research in Practice.
-See here https://worldbank.github.io/dime-data-handbook/coding.html. For more info about this linter, see https://github.com/worldbank/stata_linter.
+See here https://worldbank.github.io/dime-data-handbook/coding.html. For more info about this linter, see https://github.com/worldbank/stata-linter.
 
 ### AUTHOR:
 "DIME Analytics, DIME, The World Bank Group", [email protected]

src/lint.ado

@@ -1,4 +1,4 @@
-*! version 1.1  07dec2022  DIME Analytics [email protected]
+*! version 1.02  06apr2023  DIME Analytics [email protected]
 
 capture program drop lint
 		program 	 lint
@@ -536,7 +536,7 @@ capture program drop _checkversions
 
 	* IMPORTANT: Every time we have a package update, update the version number here
 	* Otherwise we'd be introducing a major bug!
-	local version_ado 1.1
+	local version_ado 1.02
 
 	* Check versions of .py files
 	python: from sfi import Macro

src/lint.sthlp

@@ -1,5 +1,5 @@
 {smcl}
-{* 07 Dec 2022}{...}
+{* 06 Apr 2023}{...}
 {hline}
 help for {hi:lint}
 {hline}

src/stata_linter_correct.py

@@ -1,4 +1,4 @@
-# version 1.1  07dec2022  DIME Analytics [email protected]
+# version 1.02  06apr2023  DIME Analytics [email protected]
 # Import packages ============
 import os
 import re
@@ -8,7 +8,7 @@
 # Version Global
 ## VERY IMPORTANT: Update the version number here every time there's an update
 ## in the package. Otherwise this will cause a major bug
-VERSION = "1.1"
+VERSION = "1.02"
 
 # Function to update comment delimiter =============
 # (detection works only when comment delimiter == 0)

src/stata_linter_detect.py

@@ -1,4 +1,4 @@
-# version 1.1  07dec2022  DIME Analytics [email protected]
+# version 1.02  06apr2023  DIME Analytics [email protected]
 # Import packages ====================
 import os
 import re
@@ -9,7 +9,7 @@
 # Version Global
 ## VERY IMPORTANT: Update the version number here every time there's an update
 ## in the package. Otherwise this will cause a major bug
-VERSION = "1.1"
+VERSION = "1.02"
 
 # simple run entry point
 def run():
@@ -247,11 +247,12 @@ def no_space_before_symbol(line):
 
     line = line.split('///')[0]
     groups = line.split('"')
+    pattern = r"(?:[a-z]|[A-Z]|[0-9]|_|\)|')(?:<|>|=|\+|-|\*|\^)"
 
     for i, group in enumerate(groups):
 
         if i % 2 == 0:
-            if re.search(r"(?:[a-z]|[A-Z]|[0-9]|_|\)|')(?:<|>|=|\+|-|\*|\^)", group):
+            if re.search(pattern, group):
                 return True
 
     return False
@@ -260,11 +261,12 @@ def no_space_after_symbol(line):
 
     line = line.split('///')[0]
     groups = line.split('"')
+    pattern = r"(?:(?:<|>|=|\+|-|\*|\^)(?:[a-z]|[A-Z]|_|\(|`|\.|$))|(?:(?:<|>|=|\+|\*|\^)(?:[0-9]))"
 
     for i, group in enumerate(groups):
 
         if i % 2 == 0:
-            if re.search(r"(?:<|>|=|\+|-|\*|\^)(?:[a-z]|[A-Z]|[0-9]|_|\(|`|\.|$)", group):
+            if re.search(pattern, group):
                 return True
 
     return False
@@ -413,30 +415,42 @@ def too_long_line(
     return([style_dictionary, excel_output_list])
 
 # "if" condition should be explicit
+def detect_implicit_if(line):
+
+    search_if  = re.search(r"(?:^|\s)(?:if|else if)\s", line.lstrip())
+
+    if search_if != None:
+
+        line = line[search_if.span()[0]:]
+        if (
+            (re.search(r"missing\(", line) == None) &
+            (re.search(r"inrange\(", line) == None) &
+            (re.search(r"inlist\(", line) == None) &
+            (re.search(r"=|<|>", line) == None)
+            ):
+            return True
+
+    return False
+
 def explicit_if(
     line_index, line, input_lines, indent,
     suppress, style_dictionary, excel_output_list,
     tab_space
     ):
 
     # warn if "if" statement is used but the condition is not explicit
-    search_if = re.search(r"^(if|else if) ", line.lstrip())
-    if (search_if != None):
-        if (
-            (re.search(r"missing\(", line[search_if.span()[0]:]) == None) &
-            (re.search(r"((=|<|>))", line[search_if.span()[0]:]) == None)
-            ):
-            print_output = (
-                '''Always explicitly specify the condition in the if statement. ''' +
-                '''(For example, declare "if var == 1" instead of "if var".) '''
+    if detect_implicit_if(line):
+        print_output = (
+            '''Always explicitly specify the condition in the if statement. ''' +
+            '''(For example, declare "if var == 1" instead of "if var".) '''
+            )
+        if suppress != "1":
+            print(
+                '''(line {:d}): '''.format(line_index + 1) +
+                print_output
                 )
-            if suppress != "1":
-                print(
-                    '''(line {:d}): '''.format(line_index + 1) +
-                    print_output
-                    )
-            style_dictionary["explicit_if"] += 1
-            excel_output_list.append([line_index + 1, "style", print_output])
+        style_dictionary["explicit_if"] += 1
+        excel_output_list.append([line_index + 1, "style", print_output])
 
     return([style_dictionary, excel_output_list])
 

src/stata_linter_utils.py

@@ -1,4 +1,4 @@
-# version 1.1  07dec2022  DIME Analytics [email protected]
+# version 1.02  06apr2023  DIME Analytics [email protected]
 # Import packages ====================
 import re
 import pandas as pd

stata.toc

@@ -1,3 +1,3 @@
-v 1.1
+v 1.02
 d DIME Analytics, World Bank Group, Development Economics Research
 p stata_linter

stata_linter.pkg

@@ -1,4 +1,4 @@
-v 1.1
+v 1.02
 d DIME Analytics, World Bank Group, Development Economics Research
 p stata_linter
 f /src/stata_linter_detect.py