|
1 import os |
|
2 import tempfile |
|
3 import re |
|
4 |
|
5 import pexpect |
|
6 |
|
7 from abi_error import AbiConverterError, AbiCommandError |
|
8 |
|
9 |
|
10 TYPES_IN = {'602': '602', 'abw': 'abw', 'aw': 'aw', |
|
11 'awt': 'awt', 'cwk': 'cwk', 'dbk': 'dbk', |
|
12 'doc': 'doc', 'docm': 'docm', 'docx': 'docx', |
|
13 'dot': 'dot', 'dotm': 'dotm', 'dotx': 'dotx', |
|
14 'fo': 'fo', 'htm': 'htm', 'html': 'html', |
|
15 'hwp': 'hwp', 'isc': 'isc', 'iscii': 'iscii', |
|
16 'kwd': 'kwd', 'mif': 'mif', 'odt': 'odt', |
|
17 'opml': 'opml', 'ott': 'ott', 'pdb': 'pdb', |
|
18 'pdf': 'pdf', 'rtf': 'rtf', 'sdw': 'sdw', |
|
19 'stw': 'stw', 'sxw': 'sxw', 'text': 'text', |
|
20 'txt': 'txt', 'wml': 'wml', 'wp': 'wp', |
|
21 'wpd': 'wpd', 'wri': 'wri', 'xhtml': 'xhtml', |
|
22 'xml': 'xml', 'zabw': 'zabw'} |
|
23 |
|
24 TYPES_OUT = {'abw': 'abw', 'aw': 'aw', 'awt': 'awt', |
|
25 'dbk': 'dbk', 'doc': 'doc', 'eml': 'eml', |
|
26 'fo': 'fo', 'html': 'html', 'isc': 'isc', |
|
27 'iscii': 'iscii', 'kwd': 'kwd', 'latex': 'latex', |
|
28 'mht': 'mht', 'mif': 'mif', 'nroff': 'nroff', |
|
29 'nws': 'nws', 'odt': 'odt', 'pdb': 'pdb', |
|
30 'pdf': 'pdf', 'ps': 'ps', 'rtf': 'rtf', |
|
31 'sxw': 'sxw', 'text': 'text', 'txt': 'txt', |
|
32 'wml': 'wml', 'xml': 'xml', 'xml2ps': 'xml2ps', |
|
33 'zabw': 'zabw'} |
|
34 |
|
35 class AbiFileConverter(object): |
|
36 """This let's you convert between all filetypes supperted by the |
|
37 AbiWord program. Import type isn't checked, as AbiWord doesn't check |
|
38 on extension, but on metadata. |
|
39 """ |
|
40 |
|
41 def __init__(self, timeout=60): |
|
42 self.id = None |
|
43 self.timeout = timeout |
|
44 self._start_abiword() |
|
45 |
|
46 def _start_abiword(self): |
|
47 """ |
|
48 Start abiword with the AbiCommand plugin, if not already started |
|
49 """ |
|
50 |
|
51 # find the abiword executable |
|
52 abicommand = None |
|
53 for dir in os.environ['PATH'].split(':'): |
|
54 if os.path.isfile(os.path.join(dir, 'abiword')): |
|
55 abicommand = os.path.join(dir, 'abiword') |
|
56 if not abicommand: |
|
57 raise AbiConverterError('Can not find abiword executable') |
|
58 |
|
59 # start the abiword executable |
|
60 try: |
|
61 self.child = pexpect.spawn(abicommand + ' --plugin AbiCommand') |
|
62 self.child.expect( |
|
63 'AbiWord command line plugin: Type "quit" to exit', 10) |
|
64 except: |
|
65 raise AbiConverterError('Can not open abiword executable') |
|
66 |
|
67 def stop_abiword(self): |
|
68 """ |
|
69 Stop the running abiword, kill it if necessary |
|
70 """ |
|
71 self.child.sendline('quit') |
|
72 if self._is_running(): |
|
73 os.kill(self.child.pid, 9) |
|
74 |
|
75 def _is_running(self): |
|
76 """ |
|
77 Test to see if abiword is running |
|
78 """ |
|
79 try: |
|
80 self.child.sendline('writepid /dev/null') |
|
81 self.child.expect('OK', 1) |
|
82 return True |
|
83 except: |
|
84 return False |
|
85 |
|
86 def convert_file(self, in_file, out_file=None, type=None): |
|
87 """ |
|
88 Convert a file. If out_file is not specified, a byte string is |
|
89 returned. If type is not specified, the file extension from out_file is |
|
90 used to determine the type. If this fails, the type 'text' is used. |
|
91 Return value is -1 if an error occurred. |
|
92 """ |
|
93 # is the out_file specified? |
|
94 return_bytes = False |
|
95 if out_file is None: |
|
96 out_file = tempfile.mktemp(prefix="abiconvert_") |
|
97 return_bytes = True |
|
98 |
|
99 # is the type specified |
|
100 type = TYPES_OUT.get( |
|
101 type or os.path.splitext(out_file)[1][1:], 'txt') |
|
102 |
|
103 # do the coversion |
|
104 self._perform_conversion(in_file, out_file, type) |
|
105 |
|
106 # return a byte string if no out_file is specified |
|
107 if return_bytes: |
|
108 fp = open(out_file, 'r') |
|
109 bytes = fp.read() |
|
110 fp.close() |
|
111 os.remove(out_file) |
|
112 return bytes |
|
113 |
|
114 def _perform_conversion(self, in_file, out_file, type): |
|
115 """ |
|
116 Do the actual conversion |
|
117 """ |
|
118 # make sure we are up and running |
|
119 if not self._is_running: |
|
120 self._start_abiword() |
|
121 |
|
122 # convert the file |
|
123 cmd = 'convert %s %s %s' % (os.path.abspath(in_file), |
|
124 os.path.abspath(out_file), type) |
|
125 self.child.sendline(cmd) |
|
126 |
|
127 # Check for errors |
|
128 i = self.child.expect(['OK', pexpect.TIMEOUT]) |
|
129 if i != 0: |
|
130 raise AbiCommandError('Error performing AbiCommand: %s' %cmd) |
|
131 |
|
132 def convert_to_html(self, input): |
|
133 """ |
|
134 Convert input file to HTML |
|
135 """ |
|
136 |
|
137 from tempfile import mkstemp,mkdtemp |
|
138 |
|
139 THE_OUTDIR = "outdir" |
|
140 THE_OUTFILE = "outfile" |
|
141 THE_INDIR = "indir" |
|
142 THE_INFILE = "infile" |
|
143 |
|
144 infile = None |
|
145 outfile = None |
|
146 out_f = None |
|
147 try: |
|
148 # create in/out files |
|
149 temp_dir = mkdtemp(prefix="cm_") |
|
150 |
|
151 # in |
|
152 indir_name = os.path.join(temp_dir, THE_INDIR) |
|
153 os.mkdir(indir_name) |
|
154 infile_name = os.path.join(indir_name, THE_INFILE) |
|
155 |
|
156 # out |
|
157 outdir_name = os.path.join(temp_dir, THE_OUTDIR) |
|
158 os.mkdir(outdir_name) |
|
159 outfile_name = os.path.join(outdir_name, THE_OUTFILE) |
|
160 |
|
161 # write infile |
|
162 infile = open(infile_name,'w') |
|
163 if type(input) == unicode: |
|
164 input = input.encode('utf8') |
|
165 infile.write(input) |
|
166 infile.close() |
|
167 |
|
168 # fix perms |
|
169 # TODO group permission should suffice |
|
170 os.chmod(temp_dir, 0755) # read |
|
171 os.chmod(indir_name, 0755) # read |
|
172 os.chmod(infile_name, 0755) # read |
|
173 os.chmod(outdir_name, 0777) # read / write |
|
174 |
|
175 # Do the job |
|
176 self.convert_file(infile_name, outfile_name, 'html') |
|
177 |
|
178 out_f = open(outfile_name,'r') |
|
179 output = out_f.read() |
|
180 |
|
181 # load other files (useful only for html) |
|
182 img_res = [] |
|
183 if os.path.isdir(outdir_name + '/' + THE_OUTFILE + '_files'): |
|
184 image_names = [name for name in os.listdir(outdir_name + '/' + THE_OUTFILE + '_files') if name != THE_OUTFILE] |
|
185 for image_name in image_names: |
|
186 img_res.append(os.path.join(outdir_name + '/' + THE_OUTFILE + '_files', image_name)) |
|
187 |
|
188 # clean images paths |
|
189 output = re.sub(r'<img(.+src=")outfile_files/([^"]+")', r'<img\1\2', output); |
|
190 output = re.sub(r'<img(.+)style="width:[\d\.]+mm"', r'<img\1', output); |
|
191 return output,img_res |
|
192 |
|
193 finally: |
|
194 try: |
|
195 if out_f: |
|
196 out_f.close() |
|
197 if infile: |
|
198 infile.close() |
|
199 except: |
|
200 pass |
|
201 |
|
202 def convert_from_html(self, input, format): |
|
203 """ |
|
204 Convert input file from HTML |
|
205 """ |
|
206 |
|
207 from tempfile import mkstemp,mkdtemp |
|
208 |
|
209 THE_OUTDIR = "outdir" |
|
210 THE_OUTFILE = "outfile" |
|
211 THE_INDIR = "indir" |
|
212 THE_INFILE = "infile" |
|
213 |
|
214 infile = None |
|
215 outfile = None |
|
216 out_f = None |
|
217 try: |
|
218 # create in/out files |
|
219 temp_dir = mkdtemp(prefix="cm_") |
|
220 |
|
221 # in |
|
222 indir_name = os.path.join(temp_dir, THE_INDIR) |
|
223 os.mkdir(indir_name) |
|
224 infile_name = os.path.join(indir_name, THE_INFILE + '.html') |
|
225 |
|
226 # out |
|
227 outdir_name = os.path.join(temp_dir, THE_OUTDIR) |
|
228 os.mkdir(outdir_name) |
|
229 outfile_name = os.path.join(outdir_name, THE_OUTFILE) |
|
230 |
|
231 # write infile |
|
232 infile = open(infile_name,'w') |
|
233 if type(input) == unicode: |
|
234 input = input.encode('utf8') |
|
235 infile.write(input) |
|
236 infile.close() |
|
237 |
|
238 # fix perms |
|
239 # TODO group permission should suffice |
|
240 os.chmod(temp_dir, 0755) # read |
|
241 os.chmod(indir_name, 0755) # read |
|
242 os.chmod(infile_name, 0755) # read |
|
243 os.chmod(outdir_name, 0777) # read / write |
|
244 |
|
245 # Do the job |
|
246 self.convert_file(infile_name, outfile_name, format) |
|
247 |
|
248 out_f = open(outfile_name,'r') |
|
249 output = out_f.read() |
|
250 return output |
|
251 |
|
252 finally: |
|
253 try: |
|
254 if out_f: |
|
255 out_f.close() |
|
256 if infile: |
|
257 infile.close() |
|
258 #top = temp_dir |
|
259 #for root, dirs, files in os.walk(top, topdown=False): |
|
260 # for name in files: |
|
261 # os.remove(os.path.join(root, name)) |
|
262 # for name in dirs: |
|
263 # os.rmdir(os.path.join(root, name)) |
|
264 #os.rmdir(top) |
|
265 except: |
|
266 pass |
|
267 |
|
268 def add_html_header(self, body): |
|
269 """ |
|
270 Add an HTML header to an HTML body |
|
271 """ |
|
272 |
|
273 return """ |
|
274 <html xmlns="http://www.w3.org/1999/xhtml"> |
|
275 <head> |
|
276 <meta http-equiv="content-type" content="text/html; charset=utf-8" /> |
|
277 </head> |
|
278 <body> |
|
279 %s |
|
280 </body> |
|
281 </html> |
|
282 """ %body |
|
283 |