Age | Commit message (Collapse) | Author |
|
* email/email.scm (define-cfws-pattern): Indent better.
|
|
* email/email.scm (id-left, id-right): Give higher precedence to
obsolete patterns.
|
|
* email/email.scm (obs-phrase-list, obs-utext, obs-unstruct,
obs-optional): New macros.
(unstructured, in-reply-to, references, keywords, optional-field):
Include obsolete patterns.
|
|
* email/email.scm (received): Include obsolete pattern.
(parse-mime-entity): Post process obsolete received forms.
|
|
* email/email.scm (define-atom-pattern): Do not capture cfws unless
specified.
(atom): Do not specify cfws.
(define-dot-atom-pattern): Do not capture cfws.
(define-word-pattern): New macro.
(cfws-captured-atom, cfws-captured-word): New patterns.
(obs-phrase): Use cfws-captured-word.
(received-token): Capture all.
(parse-mime-entity): Post process received and received-token.
* tests/email.scm ("parse email headers"): Fix test.
|
|
* email/email.scm (obs-day-of-week, obs-day, obs-year, obs-hour,
obs-minute, obs-second, obs-zone): New macros.
(day-of-week, day, year, hours, minutes, seconds, zone): Include
obsolete pattern.
(parse-email-headers): Handle obsolete two and three digit years, and
alphabetic time zone specifiers.
* tests/email.scm ("RFC5322 A.6.2. Obsolete dates"): New test.
|
|
* email/email.scm (obs-qp, obs-fws, obs-no-ws-ctl, obs-ctext,
obs-qtext, obs-phrase, obs-local-part, obs-dtext, obs-domain,
obs-domain-list, obs-route, obs-angle-addr, captured-atom,
captured-obs-domain, captured-domain, obs-mbox-list, obs-group-list,
obs-addr-list, obs-id-left, obs-id-right): New patterns.
(quoted-pair, fws, ctext, qtext, phrase, dtext,
define-angle-addr-pattern, mailbox-list, group-list, address-list,
define-field-pattern, from, sender, bcc, id-left, id-right,
resent-from, resent-sender, resent-bcc, obs-resent-rply): Include
obsolete pattern.
(define-printable-ascii-character-pattern-with-obsolete,
define-atom-pattern, define-obs-domain-pattern): New macros.
(define-domain-pattern): Accept obs-domain as a new argument.
(fields): Include obs-resent-rply.
* tests/email.scm ("RFC5322 A.6.1. Obsolete addressing"): New test.
("parse email addresses with period in name"): Mark as passing.
|
|
* tests/email.scm ("parse email addresses with period in name"): New
test.
Reported-by: Ricardo Wurmus <rekado@elephly.net>
|
|
We are switching from the old mailman managed mailing list to a new
public inbox. It is simpler to maintain and better suited for small
projects.
* README.org (Contributing): Switch from mailing list to public inbox.
|
|
* tests/base64.scm ("base64 random bytevector: base64-encode and
base64-decode are inverses of each other", "base64 random
bytevector: encoded output should not be more than 76 columns wide",
"base64 random bytevector: encoded output must only consist of
characters from the base64 alphabet"): Test inputs of different lengths.
* tests/quoted-printable.scm ("quoted-printable random bytevector:
quoted-printable-encode and quoted-printable-decode are inverses of
each other", "quoted-printable random bytevector: encoded output
should not be more than 76 columns wide", "quoted-printable random
bytevector: encoded output must only consist of printable ASCII
characters", "q-encoding random bytevector: q-encoding-encode and
q-encoding-decode are inverses of each other"): Test inputs of
different lengths.
|
|
The new base64 decoder can directly operate on bytevectors in addition
to strings. This feature may not remain forever, but it greatly
improves performance. So, it stays for now.
* email/email.scm (decode-body): Decode base64 encoded body directly
without converting to an intermediate string.
|
|
The new base64 decoder skips invalid characters safely.
* email/email.scm (decode-body): Do not filter base64 encoded body to
remove invalid base64 characters.
|
|
* tests/encoding.scm: Move common code for encoding tests to this new
file.
* tests/quoted-printable.scm: Load encoding.scm.
* tests/base64.scm: New file.
* Makefile.am (SCM_TESTS): Register tests/base64.scm.
|
|
* email/base64.scm: Replace file.
|
|
* email/utils.scm (bytevector-match, bytevector-overlap,
lookahead-bytevector-n): New functions.
(read-bytes-till): Do not match sequence byte by byte. Process blocks
of bytes at a time.
|
|
* email/utils.scm (not-end-let): New macro.
* .dir-locals.el (scheme-mode): Indent not-end-let correctly.
|
|
* email/utils.scm (read-while, read-bytes-till): Do not return eof if
matched at beginning. Return empty string or bytevector respectively.
* tests/utils.scm ("read-bytes-till returns empty bytevector on match
at beginning", "read-while returns empty string on match at
beginning"): New tests.
|
|
* build-website.el (org-html-head): Add minimal CSS.
|
|
* NEWS: Update news.
* README.org (Download): Link to guile-email 0.2.2 tarball.
* configure.ac: Bump to version 0.2.2.
* releases/guile-email-0.2.2.tar.lz,
releases/guile-email-0.2.2.tar.lz.asc: New files.
|
|
* tests/email.scm ("References header with only one reference must be
a singleton list, not a string"): New test.
|
|
* README.org: Advertise usage in the Guix data service.
|
|
* build-aux/test-driver.scm.in (my-gnu-runner): Log colorized test
results to stderr.
|
|
* configure.ac: Check for guile 3.0 before checking for guile 2.2.
|
|
* configure.ac: Remove custom code checking for guild.
|
|
* build-aux/test-driver.scm.in (my-gnu-runner): Add custom test-runner
on-group-begin and on-group-end functions. Accept log file port as an
argument.
|
|
* email/base64.scm: Import (rnrs arithmetic bitwise), (rnrs arithmetic
fixnums), (rnrs base), (rnrs bytevectors) and (rnrn io ports), not all
of (rnrs).
|
|
* Makefile.am (.scm.go): Use guild instead of guile-tools to build scm
files.
|
|
* tests/email.scm ("Keywords header must be a list"): Title case
header name. Remove unnecessary blank line.
|
|
* email/email.scm (post-process-fields): Treat blank Subject headers
as having the null string as value.
* tests/email.scm ("blank Subject header must be treated as having the
null string as value"): New test.
Reported-by: Ricardo Wurmus <rekado@elephly.net>
|
|
* email/email.scm (parse-email-headers): Return keywords header as a
list of strings.
* tests/email.scm ("keywords header must be a list"): New test.
|
|
* email/email.scm (body->mime-entities, email->headers+body): Reindent
calls to call-with-port.
* email/quoted-printable.scm (quoted-printable-encode,
q-encoding-encode): Reindent calls to call-with-port.
* tests/utils.scm ("read-bytes-till returns eof-object on end of
file"): Reindent call to call-with-port.
|
|
* .dir-locals.el: New file.
|
|
* email/email.scm (post-process-content-type): Use alist-combine to
override charset more strongly than just appending to the alist.
* tests/email.scm ("tolerate invalid charset"): Update test.
|
|
* email/utils.scm (alist-combine): New function.
(alist-delete*): Delete function.
* email/email.scm (add-default-headers,
add-default-mime-entity-headers): Use alist-combine.
|
|
* email/email.scm (post-process-fields): New function.
(parse-mime-entity, decode-body): Invoke post-process-fields.
|
|
* tests/email.scm (alist=?, body=?, email=?, mime-entity=?): New
functions.
(test-alist=, test-email=): New macros.
("parse email headers", "tolerate invalid charset", "tolerate decoding
errors in body"): Use test-alist=.
("RFC5322 A.1.1. A message from one person to another with simple
addressing", "parse email headers", "RFC2046 5.1.1. Common syntax",
"decode MIME entity without headers", "multipart email with a 8 bit
encoding and non UTF-8 charset part", "decode utf-8 characters in
headers", "tolerate non-ascii non-utf-8 characters in headers",
"tolerate decoding errors in body"): Use test-email=.
|
|
* email/email.scm (decode-body): Tolerate decoding errors in the body
using the substitute conversion strategy.
* tests/email.scm ("tolerate decoding errors in body"): New test.
|
|
* email/email.scm (post-process-content-type): If charset is invalid,
assume default UTF-8 as charset.
* tests/email.scm ("tolerate invalid charset"): New test.
Reported-by: Ricardo Wurmus <rekado@elephly.net>
|
|
* email/email.scm (decode-mime-encoded-word): Tolerate decoding errors
in MIME encoded words using the substitute conversion strategy.
* tests/email.scm ("tolerate decoding errors in MIME encoded words"):
New test.
Reported-by: Christopher Baines <mail@cbaines.net>
|
|
* email/email.scm (unbracketed-angle-addr): Delete duplicate
definition.
|
|
* NEWS: Update news.
* README.org (Download): Add guile-email-0.2.1 tarball.
* configure.ac: Bump version to 0.2.1.
* releases/guile-email-0.2.1.tar.lz,
releases/guile-email-0.2.1.tar.lz.asc: New files.
|
|
The earlier docstring was one meant for read-next-email-in-mbox.
* email/email.scm (mbox->emails): Update docstring.
|
|
* email/email.scm (read-next-email-in-mbox): Add docstring.
|
|
* email/email.scm (email->headers+body): If non-ASCII non-UTF-8
characters occur in the headers, do not raise a decoding error. Work
around using the substitute conversion strategy.
* tests/email.scm ("tolerate non-ASCII characters in headers"): Rename
to "decode utf-8 characters in headers".
("tolerate non-ascii non-utf-8 characters in headers"): New test.
Reported-by: Christopher Baines <mail@cbaines.net>
|
|
Guile looks for compiled go files in libdir. So, put them there.
* Makefile.am (godir): Set to libdir, instead of moddir.
|
|
* guix.scm: New file.
|
|
We tolerate non-ASCII characters in headers in order to support Emacs
message mode parens style addresses.
* email/email.scm (email->headers+body): Read headers as UTF-8
characters.
* tests/email.scm ("tolerate non-ascii characters in headers"): New
tests.
Reported-by: Christopher Baines <mail@cbaines.net>
|
|
* Makefile.am (website): Publish manual to link with version and
language information.
* README.org (Documentation): Update link to manual.
|
|
* Makefile.am (website): Publish releases folder on website.
* README.org (Download): New section.
|
|
* releases/guile-email-0.2.0.tar.lz,
releases/guile-email-0.2.0.tar.lz.asc: New files.
|