--- a/header.py Sat May 25 09:41:33 2019 +0200
+++ b/header.py Sat May 25 18:37:55 2019 +0200
@@ -1,5 +1,6 @@
import re
import linetypes
+import datetime
class Header:
def __init__(self):
@@ -10,12 +11,16 @@
self.filetype = None
self.qualifiers = None
self.license = None
- self.help = []
+ self.help = ''
self.bfc = None
self.category = None
- self.keywords = []
+ self.keywords = ''
self.cmdline = None
self.history = []
+ self.first_occurrence = dict()
+ @property
+ def valid(self):
+ return True
class BadHeader:
def __init__(self, index, reason):
@@ -27,6 +32,9 @@
index = self.index,
reason = self.reason,
)
+ @property
+ def valid(self):
+ return False
geometrical_types = [
linetypes.LineSegment,
@@ -57,6 +65,16 @@
def __str__(self):
return reason
+class HistoryEntry:
+ def __init__(self, date, user, text):
+ self.date, self.user, self.text = date, user, text
+ def __repr__(self):
+ return str.format(
+ 'HistoryEntry({date!r}, {user!r}, {text!r})',
+ date = self.date,
+ user = self.user,
+ text = self.text)
+
class HeaderParser:
def __init__(self):
self.model_body = None
@@ -64,6 +82,7 @@
self.problems = []
def parse(self, model_body):
result = Header()
+ self.result = result
self.order = []
self.cursor = -1
self.model_body = model_body
@@ -72,11 +91,11 @@
self.skip_to_next()
result.name = self.parse_pattern('^Name: (.+)$', 'name')[0]
self.skip_to_next()
- result.author = self.parse_pattern('^Author: (.+)$', 'author')[0]
+ result.author, result.username = self.parse_pattern(r'^Author: ([^\[]+)\s*(?:\[(.+)\])?$', 'author')
for header_entry in self.get_more_header_stuff():
if self.try_to_match(
'^!LDRAW_ORG ' +
- '(' \
+ r'(' \
'(?:Unofficial_)?' \
'Part|' \
'Subpart|' \
@@ -85,15 +104,48 @@
'48_Primitive|' \
'Shortcut' \
')\s?' \
- '(ORIGINAL|UPDATE \d\d\d\d-\d\d)?$',
+ '(.*)$',
'part type'):
- result.filetype, result.qualifiers = self.groups
+ result.filetype = self.groups[0]
+ result.qualifiers = re.findall(r'(?:Physical_Colour|Alias|ORIGINAL|UPDATE \d\d\d\d-\d\d)', self.groups[1])
elif self.try_to_match(
'^!LICENSE (.+)$',
'license'):
- result.license = self.groups()
+ result.license = self.groups[0]
+ elif self.try_to_match(
+ 'BFC (CERTIFY CW|CERTIFY CCW|NOCERTIFY)',
+ 'bfc'):
+ result.bfc = self.groups[0]
+ elif self.try_to_match(
+ r'!HISTORY (\d{4}-\d{2}-\d{2}) ([\[{][^\]}]+[\]}]) (.+)$',
+ 'history'):
+ result.history.append(HistoryEntry(
+ date = datetime.datetime.strptime(self.groups[0], '%Y-%m-%d').date(),
+ user = self.groups[1],
+ text = self.groups[2],
+ ))
+ elif self.try_to_match(
+ r'!HELP (.+)',
+ 'help'):
+ if result.help:
+ result.help += '\n'
+ result.help += self.groups[0]
+ elif self.try_to_match(
+ r'!CATEGORY (.+)',
+ 'category'):
+ result.category = self.groups[0]
+ elif self.try_to_match(
+ r'!KEYWORDS (.+)',
+ 'keywords'):
+ if result.keywords:
+ result.keywords += '\n'
+ result.keywords += self.groups[0]
+ elif self.try_to_match(
+ r'!CMDLINE (.+)',
+ 'cmdline'):
+ result.cmdline = self.groups[0]
else:
- self.parse_error("couldn't handle header metacommand: " + repr(header_entry.text))
+ self.parse_error("couldn't understand header syntax: " + repr(header_entry.text))
return {
'header': result,
'end-index': self.cursor + 1,
@@ -123,6 +175,7 @@
def try_to_match(self, pattern, patterntype):
try:
self.groups = self.parse_pattern(pattern, patterntype)
+ return True
except:
return False
def current(self):
@@ -133,6 +186,8 @@
match = re.search(pattern, self.current())
if match:
self.order.append(description)
+ if description not in self.result.first_occurrence:
+ self.result.first_occurrence[description] = self.cursor
return match.groups()
else:
self.parse_error(str.format("couldn't parse {}", description))