hurl 1.0

Django has nice routing, but it's too low level. Regexps are powerful, but but have cryptic syntax. hurl is a Python library that strives to make writing DRY URLs a breeze.

Consider a standard urls.py:

urlpatterns = patterns('blog.entries.views',
 url(r'^$', 'recent_entries', name='entries_recent_entries'),
 url(r'^(?P< entry_slug >[\w-]+)/$', 'show_entry', name='entries_show_entry'),
 url(r'^(?P< entry_slug >[\w-]+)/new/$', 'new_entry', name='entries_new_entry'),
 url(r'^(?P< entry_slug >[\w-]+)/edit/$', 'edit_entry', name='entries_edit_entry'),
 url(r'^(?P< entry_slug >[\w-]+)/delete/$', 'delete_entry', name='entries_delete_entry'),
 url(r'^(?P< entry_slug >[\w-]+)/comments/$', 'comments_list', name='entries_comments_list'),
 url(r'^(?P< entry_slug >[\w-]+)/comments/(\d+)/$', 'comment_details', name='entries_comment_detail'),
)

It has many issues:

- you need to remember about the '^' and the '$'
- you repeat the entry_slug url
- you need to remember arcane named group syntax
- you repeat the [\w-]+ group
- you associate name with urls conf

Better way of writing urls would be:

urlpatterns = hurl.patterns('blog.entries.views', {
 '': 'recent_entries',
 '< entry_slug:str >': {
 '': 'show_entry',
 'new': 'new_entry',
 'edit': 'edit_entry',
 'delete': 'delete_entry',
 'comments': 'comments_list',
 'comments/< :int >': 'comment_detail',
 }),
)

It conveys url structure more clearly, is much more readable and avoids repetition.

More examples

Django tutorial:

# original:
urlpatterns = patterns('',
 (r'^articles/2003/$', 'news.views.special_case_2003', {}, 'news_special_case_2003'),
 (r'^articles/(?P< year >\d{4})/$', 'news.views.year_archive', {}, 'news_year_archive'),
 (r'^articles/(?P< year >\d{4})/(?P< month >\d{2})/$', 'news.views.month_archive', {}, 'news_month_archive'),
 (r'^articles/(?P< year >\d{4})/(?P< month >\d{2})/(?P< day >\d{2})/$', 'news.views.article_detail', {}, 'news_article_detail'),
)

# hurled:
hurl = Hurl(name_prefix='news')
hurl.matchers['year'] = r'\d{4}'
hurl.matchers['month'] = r'\d{2}'
hurl.matchers['day'] = r'\d{2}'

urlpatterns = hurl.patterns('news.views', {
 'articles': {
 '2003': 'special_case_2003',
 '< year >': 'year_archive',
 '< year >/< month >': 'month_archive',
 '< year >/< month >/< day >': 'article_detail',
 }
})


Product's homepage

Requirements:

· Python