view mercurial/templater.py @ 5795:520a8d9d92ed

merge with main
author Thomas Arendsen Hein <thomas@intevation.de>
date Fri, 04 Jan 2008 07:44:07 +0100
parents 46c5e1ee8aaa
children d0576d065993
line wrap: on
line source

# templater.py - template expansion for output
#
# Copyright 2005, 2006 Matt Mackall <mpm@selenic.com>
#
# This software may be used and distributed according to the terms
# of the GNU General Public License, incorporated herein by reference.

from i18n import _
from node import *
import cgi, re, sys, os, time, urllib, util, textwrap

def parsestring(s, quoted=True):
    '''parse a string using simple c-like syntax.
    string must be in quotes if quoted is True.'''
    if quoted:
        if len(s) < 2 or s[0] != s[-1]:
            raise SyntaxError(_('unmatched quotes'))
        return s[1:-1].decode('string_escape')

    return s.decode('string_escape')

class templater(object):
    '''template expansion engine.

    template expansion works like this. a map file contains key=value
    pairs. if value is quoted, it is treated as string. otherwise, it
    is treated as name of template file.

    templater is asked to expand a key in map. it looks up key, and
    looks for strings like this: {foo}. it expands {foo} by looking up
    foo in map, and substituting it. expansion is recursive: it stops
    when there is no more {foo} to replace.

    expansion also allows formatting and filtering.

    format uses key to expand each item in list. syntax is
    {key%format}.

    filter uses function to transform value. syntax is
    {key|filter1|filter2|...}.'''

    template_re = re.compile(r"(?:(?:#(?=[\w\|%]+#))|(?:{(?=[\w\|%]+})))"
                             r"(\w+)(?:(?:%(\w+))|((?:\|\w+)*))[#}]")

    def __init__(self, mapfile, filters={}, defaults={}, cache={}):
        '''set up template engine.
        mapfile is name of file to read map definitions from.
        filters is dict of functions. each transforms a value into another.
        defaults is dict of default map definitions.'''
        self.mapfile = mapfile or 'template'
        self.cache = cache.copy()
        self.map = {}
        self.base = (mapfile and os.path.dirname(mapfile)) or ''
        self.filters = filters
        self.defaults = defaults

        if not mapfile:
            return
        i = 0
        for l in file(mapfile):
            l = l.strip()
            i += 1
            if not l or l[0] in '#;': continue
            m = re.match(r'([a-zA-Z_][a-zA-Z0-9_]*)\s*=\s*(.+)$', l)
            if m:
                key, val = m.groups()
                if val[0] in "'\"":
                    try:
                        self.cache[key] = parsestring(val)
                    except SyntaxError, inst:
                        raise SyntaxError('%s:%s: %s' %
                                          (mapfile, i, inst.args[0]))
                else:
                    self.map[key] = os.path.join(self.base, val)
            else:
                raise SyntaxError(_("%s:%s: parse error") % (mapfile, i))

    def __contains__(self, key):
        return key in self.cache or key in self.map

    def __call__(self, t, **map):
        '''perform expansion.
        t is name of map element to expand.
        map is added elements to use during expansion.'''
        if not self.cache.has_key(t):
            try:
                self.cache[t] = file(self.map[t]).read()
            except IOError, inst:
                raise IOError(inst.args[0], _('template file %s: %s') %
                              (self.map[t], inst.args[1]))
        tmpl = self.cache[t]

        while tmpl:
            m = self.template_re.search(tmpl)
            if not m:
                yield tmpl
                break

            start, end = m.span(0)
            key, format, fl = m.groups()

            if start:
                yield tmpl[:start]
            tmpl = tmpl[end:]

            if key in map:
                v = map[key]
            else:
                v = self.defaults.get(key, "")
            if callable(v):
                v = v(**map)
            if format:
                if not hasattr(v, '__iter__'):
                    raise SyntaxError(_("Error expanding '%s%s'")
                                      % (key, format))
                lm = map.copy()
                for i in v:
                    lm.update(i)
                    yield self(format, **lm)
            else:
                if fl:
                    for f in fl.split("|")[1:]:
                        v = self.filters[f](v)
                yield v

agescales = [("second", 1),
             ("minute", 60),
             ("hour", 3600),
             ("day", 3600 * 24),
             ("week", 3600 * 24 * 7),
             ("month", 3600 * 24 * 30),
             ("year", 3600 * 24 * 365)]

agescales.reverse()

def age(date):
    '''turn a (timestamp, tzoff) tuple into an age string.'''

    def plural(t, c):
        if c == 1:
            return t
        return t + "s"
    def fmt(t, c):
        return "%d %s" % (c, plural(t, c))

    now = time.time()
    then = date[0]
    delta = max(1, int(now - then))

    for t, s in agescales:
        n = delta / s
        if n >= 2 or s == 1:
            return fmt(t, n)

def stringify(thing):
    '''turn nested template iterator into string.'''
    if hasattr(thing, '__iter__'):
        return "".join([stringify(t) for t in thing if t is not None])
    return str(thing)

para_re = None
space_re = None

def fill(text, width):
    '''fill many paragraphs.'''
    global para_re, space_re
    if para_re is None:
        para_re = re.compile('(\n\n|\n\\s*[-*]\\s*)', re.M)
        space_re = re.compile(r'  +')

    def findparas():
        start = 0
        while True:
            m = para_re.search(text, start)
            if not m:
                w = len(text)
                while w > start and text[w-1].isspace(): w -= 1
                yield text[start:w], text[w:]
                break
            yield text[start:m.start(0)], m.group(1)
            start = m.end(1)

    return "".join([space_re.sub(' ', textwrap.fill(para, width)) + rest
                    for para, rest in findparas()])

def firstline(text):
    '''return the first line of text'''
    try:
        return text.splitlines(1)[0].rstrip('\r\n')
    except IndexError:
        return ''

def isodate(date):
    '''turn a (timestamp, tzoff) tuple into an iso 8631 date and time.'''
    return util.datestr(date, format='%Y-%m-%d %H:%M')

def hgdate(date):
    '''turn a (timestamp, tzoff) tuple into an hg cset timestamp.'''
    return "%d %d" % date

def nl2br(text):
    '''replace raw newlines with xhtml line breaks.'''
    return text.replace('\n', '<br/>\n')

def obfuscate(text):
    text = unicode(text, util._encoding, 'replace')
    return ''.join(['&#%d;' % ord(c) for c in text])

def domain(author):
    '''get domain of author, or empty string if none.'''
    f = author.find('@')
    if f == -1: return ''
    author = author[f+1:]
    f = author.find('>')
    if f >= 0: author = author[:f]
    return author

def email(author):
    '''get email of author.'''
    r = author.find('>')
    if r == -1: r = None
    return author[author.find('<')+1:r]

def person(author):
    '''get name of author, or else username.'''
    f = author.find('<')
    if f == -1: return util.shortuser(author)
    return author[:f].rstrip()

def shortdate(date):
    '''turn (timestamp, tzoff) tuple into iso 8631 date.'''
    return util.datestr(date, format='%Y-%m-%d', timezone=False)

def indent(text, prefix):
    '''indent each non-empty line of text after first with prefix.'''
    lines = text.splitlines()
    num_lines = len(lines)
    def indenter():
        for i in xrange(num_lines):
            l = lines[i]
            if i and l.strip():
                yield prefix
            yield l
            if i < num_lines - 1 or text.endswith('\n'):
                yield '\n'
    return "".join(indenter())

def permissions(flags):
    if "l" in flags:
        return "lrwxrwxrwx"
    if "x" in flags:
        return "-rwxr-xr-x"
    return "-rw-r--r--"

common_filters = {
    "addbreaks": nl2br,
    "basename": os.path.basename,
    "age": age,
    "date": lambda x: util.datestr(x),
    "domain": domain,
    "email": email,
    "escape": lambda x: cgi.escape(x, True),
    "fill68": lambda x: fill(x, width=68),
    "fill76": lambda x: fill(x, width=76),
    "firstline": firstline,
    "tabindent": lambda x: indent(x, '\t'),
    "hgdate": hgdate,
    "isodate": isodate,
    "obfuscate": obfuscate,
    "permissions": permissions,
    "person": person,
    "rfc822date": lambda x: util.datestr(x, "%a, %d %b %Y %H:%M:%S"),
    "rfc3339date": lambda x: util.datestr(x, "%Y-%m-%dT%H:%M:%S", True, "%+03d:%02d"),
    "short": lambda x: x[:12],
    "shortdate": shortdate,
    "stringify": stringify,
    "strip": lambda x: x.strip(),
    "urlescape": lambda x: urllib.quote(x),
    "user": lambda x: util.shortuser(x),
    "stringescape": lambda x: x.encode('string_escape'),
    }

def templatepath(name=None):
    '''return location of template file or directory (if no name).
    returns None if not found.'''

    # executable version (py2exe) doesn't support __file__
    if hasattr(sys, 'frozen'):
        module = sys.executable
    else:
        module = __file__
    for f in 'templates', '../templates':
        fl = f.split('/')
        if name: fl.append(name)
        p = os.path.join(os.path.dirname(module), *fl)
        if (name and os.path.exists(p)) or os.path.isdir(p):
            return os.path.normpath(p)