forked from danizen/PyZ3950
-
Notifications
You must be signed in to change notification settings - Fork 2
/
CHANGELOG
157 lines (123 loc) · 5.7 KB
/
CHANGELOG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
2.04 (2004 Jul 20)
Added strict flag to zmarc.MARC to turn off assertions about charset,
etc. (some servers return broken MARC)
Added zmarc.MARC8_to_Unicode class to handle MARC8 to Unicode charset
conversion.
Fixed thread-safety bug which would cause multithreaded users of ccl
to mysteriously get lexing errors. Added code to report more info
about lexing errors.
Added an example Z39.50 server for use with twisted to example/twisted/*.
2.02 (2004 Jun 07)
Removed code to generate parsetab at install time (too difficult
to maintain across platforms, 2.01 was broken on Windows)
2.01 (2004 Jun 04)
Fixed bug in installation under Python 2.3 (couldn't find ccl
to generate parsetab at install time)
Fixed asn1.py to be backward compatible w/Python 2.1
2.0
[Note: major rewrites, no fix-by-fix description available.]
Fix spelling of "license" in setup.py to be US, not UK, so distutils
recognizes it.
OPACRecord attribute "bibliographicRecord" had a leading space, so
it was inaccessible with normal attribute access notation. Fixed.
0.96 (2003 May 12)
According to the spec, it's legitimate to return fewer records than
requested by a Present request. Previously, the code just returned a
null record when we made a request to fill our local cache, and
the Present response didn't populate all the cache we asked for,
and one of the unpopulated records was requested.
Bill Oldroyd found a case in which this happened because the
target had a small maximum message size. Fixed to re-request
the individual record. (added test_extract.py to test this.)
Added z3950.msg_size to set message size, to help provoke this
failure mode for testing.
0.95 (2003 Jan 23)
switch from "import <foo>" to "from PyZ3950 import <foo>" to make
pydoc work.
XML stuff from Robert Sanderson in zmarc.py
add compiler for ASN.1 text (still alpha-y)
Add stab at character set negotiation (turned off by default)
base-128 encoding for OIDs can have 0 in the middle: add support
changes for handling of ASN.1 ANY
In InitializeResponse, userInformationField was mistakenly not EXPLICIT,
so we'd barf whenever it was provided. Affects Innovatve Interfaces servers.
Thanks to Brian Kennison, kennisonb@wcsu.ctstateu.edu, for providing a test
case.
Fix zmarc_example to use os.path.join instead of '/'
Change scan interface to be a sequence of dicts
0.94 (2002 July 25)
rework authetication code a little (change to be more consistent
with rest of model, change "auth" keyword to "authentication" for
stylistic consistency, allow passing in (user,) or (user, pass) or
(user, pass, group)).
cosmetic fixes to zmarc code (no user-visible change, I hope)
Better handling of nonsurrogate diagnostics: we wouldn't always
raise the appropriate (or, in fact, any) exception. Note that
for multiple nonsurrogate diags, as per ZOOM mailing list discussion,
we only raise the first.
Better handling of surrogate diagnostics: we used to raise a
ProtocolError when we received the diagnostic, but now we raise
a Bib1Err (or other error, if the OID is different) on access to
that record.
New in 0.93 (2002 July 24)
zoom, z3950: add scan, user authentication (see zoom docstring for,
um, docs)
z3950: change implementationName passed in Initialize Request
asn1: add trace_int flag for tracing reading ints
Fix nasty bug: would sometimes raise an IndexError if we tried to peek
at the next byte, but the buffer was empty and needed to be refilled.
Fix trace_seq_of to work w/ CHOICE instead of complaining about no tag
added test/testscan.py, to demo/test scan code.
zmarc.py: add code to write out MARC data, slightly better error
reporting on read.
New in 0.92 (2002 July 22)
zoom: Fix to throw a zoom.ClientNotImpl error when a preferredRecordSyntax
we don't recognize is passed in. (Used to throw a KeyError.)
For command-line client, specify elementSetName 'F' by default, allow
overriding with -a cmd-line parameter.
z3950: support decoding SearchInfoReport (so we can talk to www.bibhit.dk,
among others: used to just throw BER exn). Added OVID db to table.
example/{zthes1, gthes}: many fixes.
New in 0.91 (2002 July 17)
General:
We now use a Python package, PyZ3950, instead of cluttering up
the top-level Python module namespace. (Sorry, this means you have
to change your import statements, but I think everyone's going
to be happier in the long run.)
asn1:
add trace_seq_of flag for debugging
add GeneralizedTime type
add facility for registering OID/ASN.1 combinations, so that
ANY types within an EXTERNAL can be properly decoded.
change BitStringVal so that out-of-range bits are returned as 0
change "Bad length for NULL" exception text
z3950:
add oid definitions for Utility, Cross-Domain-1, and Zthes-1 attribute sets,
XML and GRS-1 record formats, Zthes schema, and AuthorityFileInfo_External.
fix ASN.1 for OtherInformation (category wasn't optional)
add GRS-1 and OPAC ASN.1 definitions.
If InitializeResponse support flags for search or present aren't set, go
ahead and do the search/present anyway, instead of returning None
Remove 'print "Delete not supported"' when delete is called but
not supported.
Add some more servers to table.
bib1msg:
fix stupid bugs which would cause (wrong) exceptions to be thrown while
trying to throw a Bib1Err
grs1:
first check-in
zoom:
change docstring for Query.__init__
We used to throw an exception when taking a slice of an empty result
set. Fix to correctly return []
Support for GRS-1, XML, and OPAC record syntaxes (OPAC data is just
struct output from BER decoder).
New -f formats option for test mode.
ccl:
Add top-level ATTRSET(<attrset>/query) kludge to specify attrsets
Allow qualifiers to be literal (<numeric type>,<numeric value>) pairs.
Support relops other than just '='
Better prettyprinting of ASTs
zmarc:
added docstring.
0.90 - Start of Changelog.