Lib/textwrap.py (part 3)

Source:

cpython 3.14 @ ab2d84fe1023/Lib/textwrap.py ↗

This annotation covers the core wrapping algorithm and utility functions. See modules_textwrap2_detail for TextWrapper.__init__ and wrap.

Map

Lines	Symbol	Role
1-80	TextWrapper._wrap_chunks	Place word chunks onto lines
81-160	textwrap.dedent	Remove common leading whitespace
161-240	textwrap.indent	Add a prefix to each line
241-400	textwrap.shorten	Truncate to a max width with ellipsis

Reading

TextWrapper._wrap_chunks

# CPython: Lib/textwrap.py:320 _wrap_chunks
def _wrap_chunks(self, chunks):
    """Wrap a sequence of text chunks (alternating whitespace and words)."""
    lines = []
    chunks.reverse()  # Will pop from end (efficient)
    while chunks:
        cur_line = []
        cur_len = 0
        if lines:
            indent = self.subsequent_indent
        else:
            indent = self.initial_indent
        width = self.width - len(indent)
        
        while chunks:
            l = len(chunks[-1])
            if cur_len + l <= width:
                cur_len += l
                cur_line.append(chunks.pop())
            else:
                break
        
        if chunks and len(chunks[-1]) > width:
            self._handle_long_word(chunks, cur_line, cur_len, width)
        ...
        lines.append(indent + ''.join(cur_line))
    return lines

The algorithm greedily fills each line with as many chunks as fit. chunks alternates between whitespace and words. Reversing and popping is O(1) per chunk.

textwrap.dedent

# CPython: Lib/textwrap.py:420 dedent
def dedent(text):
    """Remove any common leading whitespace from all lines in text."""
    _whitespace_only_re = re.compile('^[ \t]+$', re.MULTILINE)
    _leading_whitespace_re = re.compile('(^[ \t]*)(?:[^ \t\n])', re.MULTILINE)
    text = _whitespace_only_re.sub('', text)
    indents = _leading_whitespace_re.findall(text)
    margin = reduce(commonprefix, indents) if indents else ''
    ...
    return re.sub(r'(?m)^' + margin, '', text)

textwrap.dedent strips the common prefix from all non-empty lines. It is idempotent. Used with triple-quoted docstrings: textwrap.dedent(''' line1\n line2''') removes the leading spaces.

textwrap.shorten

# CPython: Lib/textwrap.py:480 shorten
def shorten(text, width, **kwargs):
    """Collapse and truncate the given text to fit in the given width."""
    w = TextWrapper(width=width, max_lines=1, **kwargs)
    return w.fill(' '.join(text.split()))

textwrap.shorten('Hello World', width=8) returns 'Hello...' (using the default placeholder '...'). First normalizes whitespace, then wraps to one line with max_lines=1.

gopy notes

TextWrapper._wrap_chunks is module/textwrap.WrapChunks in module/textwrap/module.go. It works on a []string slice and uses Go's strings.Builder to assemble lines. dedent uses regexp.FindAllString to find common indentation. shorten calls strings.Fields to normalize whitespace then calls Wrap.