Add script to check Authors.rst

.travis.yml

@@ -16,6 +16,7 @@ script:
 - sh travis/git-info.sh
 - python travis/pr-check-base.py
 - python travis/lint.py
+- python travis/authors-rst.py
 - python travis/script-in-readme.py
 - python travis/script-syntax.py --ext=lua --cmd="luac5.2 -p"
 - python travis/script-syntax.py --ext=rb --cmd="ruby -c" --path scripts/

travis/authors-rst.py

@@ -0,0 +1,67 @@
+""" Overly-complicated script to check formatting/sorting in Authors.rst """
+
+import re, sys
+
+def main():
+    success = [True]
+    def error(line, msg, **kwargs):
+        info = ''
+        for k in kwargs:
+            info += ' %s %s:' % (k, kwargs[k])
+        print('line %i:%s %s' % (line, info, msg))
+        success[0] = False
+    with open('docs/Authors.rst', 'rb') as f:
+        lines = list(map(lambda line: line.decode('utf8').replace('\n', ''), f.readlines()))
+
+        if lines[1].startswith('='):
+            if len(lines[0]) != len(lines[1]):
+                error(2, 'Length of header does not match underline')
+            if lines[1].replace('=', ''):
+                error(2, 'Invalid header')
+
+        first_div_index = list(filter(lambda pair: pair[1].startswith('==='), enumerate(lines[2:])))[0][0] + 2
+        first_div = lines[first_div_index]
+        div_indices = []
+        for i, line in enumerate(lines[first_div_index:]):
+            line_number = i + first_div_index + 1
+            if '\t' in line:
+                error(line_number, 'contains tabs')
+            if line.startswith('==='):
+                div_indices.append(i + first_div_index)
+                if not re.match(r'^=+( =+)+$', line):
+                    error(line_number, 'bad table divider')
+                if line != lines[first_div_index]:
+                    error(line_number, 'malformed table divider')
+        if len(div_indices) < 3:
+            error(len(lines), 'missing table divider(s)')
+        for i in div_indices[3:]:
+            error(i + 1, 'extra table divider')
+
+        col_ranges = []
+        i = 0
+        while True:
+            j = first_div.find(' ', i)
+            col_ranges.append(slice(i, j if j > 0 else None))
+            if j == -1:
+                break
+            i = j + 1
+
+        for i, line in enumerate(lines[div_indices[1] + 1:div_indices[2]]):
+            line_number = i + div_indices[1] + 2
+            for c, col in enumerate(col_ranges):
+                cell = line[col]
+                if cell.startswith(' '):
+                    error(line_number, 'text does not start in correct location', column=c+1)
+                # check for text extending into next column if this isn't the last column
+                if col.stop is not None and col.stop < len(line) and line[col.stop] != ' ':
+                    error(line_number, 'text extends into next column', column=c+1)
+            if i > 0:
+                prev_line = lines[div_indices[1] + i]
+                if line.lower()[col_ranges[0]] < prev_line.lower()[col_ranges[0]]:
+                    error(line_number, 'not sorted: should come before line %i ("%s" before "%s")' %
+                        (line_number - 1, line[col_ranges[0]].rstrip(' '), prev_line[col_ranges[0]].rstrip(' ')))
+
+    return success[0]
+
+if __name__ == '__main__':
+    sys.exit(int(not main()))