Martin Tarenskeen writes:
> Hi again,
>
> To answer my own question: Something like
>
> {
> << { c1 } \\ { s2.\sustainOn s4\sustainOff } >>
> \bar "|."
> }
>
> seems to do the trick for me.
Why the \\ in here? It would seem that separ
odule (scm accreg) is not present in any header.
> This holds for other modules as well.
Well, the question would be what makes \discant known to be a music
function but not known to be a markup command.
Markups are weird.
--
David Kastrup
Jonas Hahnfeld writes:
> Am Sonntag, dem 24.01.2021 um 17:31 +0100 schrieb David Kastrup:
>> Jonas Hahnfeld writes:
>>
>> > Am Sonntag, dem 24.01.2021 um 16:52 +0100 schrieb David Kastrup:
>> > > Jonas Hahnfeld writes:
>> > >
>> &
Jonas Hahnfeld writes:
> Am Sonntag, dem 24.01.2021 um 16:52 +0100 schrieb David Kastrup:
>> Jonas Hahnfeld writes:
>>
>> > Am Sonntag, dem 24.01.2021 um 00:54 +0100 schrieb Valentin Petzel:
>> > > Hello,
>> > >
>> > > Lilypond tr
ave no idea why they would have changed that.
--
David Kastrup
Richard Shann writes:
> On Mon, 2021-01-04 at 00:45 +0100, David Kastrup wrote:
>> Richard Shann writes:
>>
>> > My understanding of the \beamExceptions function in 2.20 was that
>> > you
>> > should be able to pass it a complete bar beamed as you
ion "2.20"
>
> barOne = {g'16 b' b'[( a'32 g']) d'16[ c''16] c''16[( b'32 a'])}
Beaming for 1/16 notes ends here after c''16] and nowhere else. Beaming
for 32nd notes ends after g'32]) and a'32]) and nowhere else. Those are
the beaming exceptions being generated by
\beamExceptions \barOne
as you can verify with
\void \displayScheme \beamExceptions \barOne
--
David Kastrup
er than the timestep in question. That
makes unexpected side effects much less likely.
--
David Kastrup
haracters or tabs or
combining characters into account.
--
David Kastrup
ers would be
> all over it.
Indeed. And a unifying concept covering both grobs and MIDI objects
(mobs?) might even pave a way to iterative generation of time-based
MusicXML (xobs?) or Braille (bobs?) or other renditions of the musical
content.
--
David Kastrup
yPond is mainly intended for "proofhearing", not
as a performance; it has been fairly recently only that attempts to take
into account articulation marks (which are seminal for a performance)
have been added. As such, unfolding repeats makes the material longer
without adding more information.
--
David Kastrup
Jonas Hahnfeld writes:
> Am Dienstag, den 20.10.2020, 18:26 +0200 schrieb David Kastrup:
>> Jonas Hahnfeld writes:
>>
>> > I don't want to digress into this topic right now (P.S. the reply got
>> > longer than I initially anticipated), but the script
-user. This must happen in a proper thread
> on the -devel list, and hopefully with more technical content than
> just the statement of "we need something better".
--
David Kastrup
brain
surgeon.
And make no mistake: TeX/LaTeX have a much more widespread user base
than LilyPond has. So in terms of expanding the user base and its
comfort level, looking at high-level formats like LaTeX and formalising
interfaces to make it possible to build applications and extensions
independently (working with LilyPond is more akin to working with plain
TeX rather than LaTeX) seems like a more promising course.
--
David Kastrup
commit 864ce06c432a62191773b0d59401348c9b51568c
Author: David Kastrup
Date: Mon Oct 8 17:52:36 2012 +0200
Allow property and grob paths to be constructed from strings
This allows use of either
\override Accidental color = #red
or
\override Accidental.color = #red
A = "Something"
>>
>> \markup {
>>\print-if-defined #'symA "Text"
>>\print-if-defined #'symB "More text"
>> }
>>
>
> An empty stencil will still be spaced (unless removed by other
> markup-commands).
empty-stencil will typically not trigger additional spacing. That's
typically what distinguishes it from point-stencil .
--
David Kastrup
nse to add crucial information
to them, even if the likelihood that the original participants care is
pretty much zero.
Admittedly, it happens more to discussions concerning vintage digital
cameras, but there have been two or three occasions where I even did it
with LilyPond. Typically somewhat facetiously when some long-requested
feature or bug had finally been addressed, probably in the wake of
structural changes making it possible in the first place.
--
David Kastrup
only way in which you can arrive at a
redistributable whole.
The main problem of adding _anything_ to a license of the GPL/LGPL
flavor is that the result is incompatible to software licensed under
GPL/LGPL without such an addition, defeating the significant goal of
interoperability and reusability of the GNU code base.
--
David Kastrup
frescobaldi-lis_ksvl/tmpgdr3w9ef/J-Haydn-Minuet in
> C-correct.ly:14:1 <0>: error: not a note name: fz
>
>
> fz = #(make-dynamic-script "fz")
Well, apparently line 14 in your file is not a good place. Since you
don't show any other line, it's hard to say anything just why it is not
a good place.
--
David Kastrup
s not a technical one; it's that they are simply not useful or not
> useful enough on an organ.
I've seen a piece where an organist fixed keys in place with some kind
of wedge that seemed designed for that purpose.
--
David Kastrup
quot; is, after
> all, and the fact that it isn't mentioned in Section 1.3 tends to make
> it look as if Lilypond just isn't capable of "rit." - which of course
> seems strange. A person who's at the stage of trying to find out how
> to put a "rit." into their score is unlikely to know that they need to
> look in Section 1.8.
Uh, \tempo "rit." ?
--
David Kastrup
ed with non-GPL
licensed software (such as Dorico) as long as you provided the source
code of your OLL derivative in a manner where people could change and
relink with changed versions.
That's more or less all. I haven't seen anybody wanting to do anything
of that kind, though.
--
David Kastrup
n, you would win every trial
> in every legal area which accepted the GPL as an effective
> license. That's the risk I would have to take, if I used OpenLilyLib
> to ease my work.
That's just garbage, sorry. It would apply for LSR snippets which are
not used via an interface but are code pieces to be copied and adapted
into your source. For LSR code, a GPL license would be quite
problematic (by the way, what is the licensensing of LSR pieces?). But
that is not how OLL is used.
--
David Kastrup
particular version of OLL code, this source code and
its derivatives (presumably including the generated PDF) can only be
distributed under the GPL.
That's not the intended way of using OLL, of course.
--
David Kastrup
cglyph "scripts.caesura.straight"
> }
> \breathe
> c''
> \breathe
> }
>
> Cheers,
> Kieren.
> ____
>
> Kieren MacMillan, composer (he/him/his)
> ‣ website: www.kierenmacmillan.info
> ‣ email: kie...@kierenmacmillan.info
>
>
>
--
David Kastrup
ously call
#(use-modules (scm display-lily))
--
David Kastrup
16 d2. }
\new Staff \relative { c''4 \bar ".|:" \grace s16 d2. }
>>
Please make sure that you use the ‘\grace’ command for the spacer
part, even if the visual part uses ‘\acciaccatura’ or ‘\appoggiatura’
because otherwise an ugly slur fragment will be printed, connecting the
invisible grace note with the following note.
--
David Kastrup
not to bother a lot with keeping rests around or keeping a
two-voiced look.
So your change might have been the better choice.
--
David Kastrup
cros be of very limited usefulness for processing
expressions specified as music function arguments.
--
David Kastrup
David Kastrup writes:
> Maurits Lamers writes:
>
>> Hi,
>>
>>> convert-ly does text replacements. It is not a full parser. If text
>>> replacements are supposed to work, you need to write your text in a way
>>> that the replacement patt
's own style, but things like # or
#( are so unusual that they have not made it into the
patterns.
--
David Kastrup
ite your text in a way
that the replacement patterns cover. Stuff like putting # on one line
and a corresponding opening paren on the next line are just too weird
for those writing the conversion rules to have foreseen.
So first try formatting your source in a somewhat common manner and then
try running convert-ly.
--
David Kastrup
y of simplifying
> this? Ideally, it'd be nice to be able to write something like
>
> f4.^{\mx 1.5 -1.5}
>
> and leave all the once override stuff out of the score itself. Or is there
> another way of placing a symbol where I want it?
Why wouldn't you use a tweak rather than an override?
--
David Kastrup
Aaron Hill writes:
> On 2020-08-29 6:44 am, David Kastrup wrote:
>> Hm. \with instead of = would even fit into the parser. But that
>> leaves
>> tweaks in the lurch.
>
> Not sure I am following. Are you indicating that something like...
>
>
> \onc
gt; number-type = #'arabic
> orientation = #'landscape
> mute-string = #"M"
> label-dir = #LEFT
> dot-color = #'black
> }
>
Hm. \with instead of = would even fit into the parser. But that leaves
tweaks in the lurch.
--
David Kastrup
Aaron Hill writes:
> On 2020-08-29 5:23 am, David Kastrup wrote:
>> Wols Lists writes:
>>
>>> On 29/08/20 05:45, Werner LEMBERG wrote:
>>>> \once \override FretBoard.size = #'1.0
>>>> \once \override FretBoard.fret-diagram-detai
ard.fret-diagram-details =
#'((barre-type . straight)
(dot-color . black)
(finger-code . below-string))
I mean, I am as proud as anybody that separate subproperty overrides
don't cause LilyPond to go down in flames any more, but this is an
alist, after all.
--
David Kastrup
ng
FretBoard
So any commands stacked before this last \propertyTweak command have no
way of knowing that they are supposed to assume fret-diagram-details as
a given.
>> And actually using this syntax for the final \propertyTweak is
>> likely going to be a lot more confusing than just staying like you
>> are.
>
> I agree. It was a bad example, and the one above should better
> demonstrate what I'm looking for.
No idea what you mean.
--
David Kastrup
tacking
_multiple_ \propertyTweak commands, so it doesn't help a lot in your
case. And actually using this syntax for the final \propertyTweak is
likely going to be a lot more confusing than just staying like you are.
--
David Kastrup
think
> that wrapping is suboptimal and reduces legibility I hope you have
> some trust in my decision...
Maybe
\void \displayLilyMusic
\once
\propertyTweak color #red
\propertyTweak font-size #3
\propertyTweak direction #UP Voice.Slur
helps?
--
David Kastrup
line.
> Theoretically, the snippet could be printed with a smaller font size,
> but this doesn't look very pretty IMHO. I consider the `\oo`
> shorthand both innocuous and simple enough for a snippet.
I find a line wrap quite less of a distraction.
--
David Kastrup
in itself tries is about showcasing the
shorthand. It detracts from the content.
--
David Kastrup
specific enough to lead to error
messages at the place where the wrong input is rather than triggering
followup errors.
Given correct input, they should do the same.
--
David Kastrup
you can aim for is
>>
>> \oo foo.bar #'baz
>
> This would be just fine. The thing is to replace `\once\override`
> with something shorter.
oo = #(define-music-function (parser location prop value) (symbol-list? scheme?)
#{ \once \override #prop = #value #})
should likely work fine in 2.18.
--
David Kastrup
ng, like letting oo inject "\\once \\override "
into the input stream, it's very likely that it will break down in more
complex uses than when just writing it into immediate code.
> Help would be much appreciated.
--
David Kastrup
parts after the tagged section.
> However, this can be got around by adding a \partCombineChords
> instruction before the 3rd beat of the bar.
>
> Does anyone know of a way of correcting the extended staff?
>
> David
>
>
>
--
David Kastrup
ited to 32bit, has much worse programming features and has no
useful online support by current users and programmers because everyone
has moved on.
To get anywhere useful except for a single-shot project, you'll not get
around upgrading, and upgrading may make getting anywhere useful easier.
Just saying.
--
David Kastrup
eased, and is
> now at the same time much more feature-rich _and_ easier to use.
Well, that's not much of a motivation to upgrade existing documents
(unless you want to add to them). But the typesetting has become quite
a bit better, too.
--
David Kastrup
nds on and that is lilypond 2.14. I am afraid that it
> would distract the current effort and cost too much time at the moment
> to convert that body of work to a more recent version of lilypond.
Have you even tried running convert-ly on it?
--
David Kastrup
up_identifier
("bookpart-text-handler");
lily/parser.yy: SCM proc = parser->lexer_->lookup_identifier
("bookpart-score-handler");
lily/parser.yy: ("output-def-music-handler");
lily/parser.yy: ("output-def-music-handler");
lily/parser.yy: SCM proc = parser->lexer_->lookup_identifier
("context-mod-music-handler");
lily/parser.yy: SCM proc = parser->lexer_->lookup_identifier
("context-mod-music-handler");
--
David Kastrup
Paul McKay writes:
>> On Tue, 11 Aug 2020, 17:53 David Kastrup, wrote:
>>
>>> Paul McKay writes:
>>>
>>> > Hi Aaron.
>>> > Many thanks for this. I can now write functions like this:
>>> > mock = #(define-music-function
>>&
- eth at us } \etc
--
David Kastrup
ad checks whether it can interpret \mynote in a manner more
acceptible to \displayMusic. When it finds a valid interpretation
passing the predicate, it ultimately calls displayMusic with it.
--
David Kastrup
the grob.
I think most of the time the handwaving heuristic "let a line break mark
all information not corresponding to the current key signature as
unknown" would work with most accidental rules.
--
David Kastrup
o cater in advance for
all possible line breaks and how to write rules in anticipation of it.
--
David Kastrup
led that it can access.
> No pre-resampling cache hit
> Last 1 MIDI events are ignored
> Playing time: ~4 seconds
> Notes cut: 0
> Notes lost totally: 0
>
> When I try to read my midi file. Can someone help me ?
I think your Midi file is likely fine. You just don't have what it
takes to play it.
--
David Kastrup
Stanton Sanderson writes:
>> On Aug 7, 2020, at 3:31 PM, David Kastrup wrote:
>>
>> Stanton Sanderson writes:
>>
>>> I have had no response to my previous post, so I will try again.
>>>
>>> On Mac OS 10.14.6 (Mojave), running LilyPond
in
> sys.stdin = codecs.getreader('utf8')(sys.stdin.detach())
> AttributeError: 'file' object has no attribute ‘detach'
>
>
> I saw the same behavior with 2.21.1 and 2.21.3 when I attempted to update
> older files.
>
> What am I missing?
A recent enough version of Python? Can you figure out what Python
version is used by your installation?
--
David Kastrup
ke-volta-set make-volta-set)
(define (make-volta-set ...) ...)
Possibly
(set! make-volta-set (lambda (...) ...))
but I think that the define probably works.
>
> Thank you for your help; have a great Summer!
> — Niols
>
>
--
David Kastrup
David Kastrup writes:
> Thomas Morley writes:
>
>> Hi,
>>
>> the following code generates outlined nested pdf-marks, see attached.
>>
>> \version "2.21.4"
>>
>> \book {
>> \markuplist \table-of-contents
>> \tocItem x
}
> }
>
> Though, if I switch from \book to \bookpart pdf-marks are gone (while
> the table-of-content is still unaffected).
Wouldn't that warrant putting the \bookpart inside of a \book? That's
sort of it's point, isn't it?
--
David Kastrup
Niols writes:
> Hi David,
>
> On 29/07/2020 15:16, David Kastrup wrote:
>> Niols writes:
>>> Must I understand that it is not possible to write iterators in Scheme
>>> and that they have to be written in C++?
>> Yes.
>
> Thank you; at le
onal use, I would have to recompile LilyPond myself?
Yes, or reuse one of the existing ones. Some are comparatively generic.
> If it is actually possible to write an iterator in Scheme, would
> anyone have an example to show me or a link to give me?
>
> Thank you so much in advance.
What do you want to achieve?
--
David Kastrup
David Kastrup writes:
> Aaron Hill writes:
>
>> On 2020-07-28 10:15 am, David Kastrup wrote:
>>> Aaron Hill writes:
>>>
>>>> Feels like a hack, but would this help?
>>>>
>>>> \version "2.20.0"
>>>>
Aaron Hill writes:
> On 2020-07-28 10:15 am, David Kastrup wrote:
>> Aaron Hill writes:
>>
>>> Feels like a hack, but would this help?
>>>
>>> \version "2.20.0"
>>> loremIpsum = \markuplist {
>>> \bold { L
(line-width . 40)
> \wordwrap { $@loremIpsum $@loremIpsum $@loremIpsum }
> }
>
Why $@loremIpsum rather than \loremIpsum ?
--
David Kastrup
lude file; }
\markup {
\column { \override #'(line-width . 92) {
\wordwrap {
\PrayerBeginning
\TodaysMiddle
\PrayerEnding
}
} }
}
--
David Kastrup
tch-alteration #{ck#}))
--
David Kastrup
that
> allows me to switch between those three options (dotted quarter,
> quarter with tenuto, or half note) relatively easily.
That would then be reasonably simple.
--
David Kastrup
t could be going wrong.
My money would be on issue 34. Namely parallel context-starting events
for which some have grace timing and some not.
--
David Kastrup
nd's Guile interfaces now make sure that the value
actually can be represented by the given type. The value here would
appear to be -1 which has no representation in an unsigned int.
No idea where this happens, but maybe this helps someone else figuring
out where OpenLily was likely previously triggering undefined behavior
and now causes an error exit.
--
David Kastrup
Urs Liska writes:
> Am Dienstag, den 14.07.2020, 22:57 +0200 schrieb David Kastrup:
>>
>> That is an incorrect description. This only happens when you comment
>> out only the first value. When you comment out both optional values,
>> the first value that is see
discourage messing with that part of the design.
It has taken a long time (basically from 2.14 to 2.20) to make all
aspects of this design work dependably, not just in terms of it working
in a logical manner and as described, but also in terms of being
predictable and matching a large set of use cases (a whole lot of what
now is implemented in terms of music functions was hardwired syntax
previously).
Changes of the "I know, I'll just change the parser semantics for that
particular use case I care about" kind will very likely lead to a
cascading sequence of consequences that have no logical (let alone
programmatic) solution and will be discovered by and by.
--
David Kastrup
standard 4 part
quartet scores differently) before being even allowed to write the first
line actually tackling the problem tends to make a difference to me.
> The question was to ask if it can be done.
It wasn't clear to me that a simple "yes" would have satisfied you.
--
David Kastrup
he complete puzzle on your own, possibly just to
hear afterwards "thanks, but that is not my actual problem" or "thanks,
this was not really important to me and I decided to do something else"
or "thanks, but that approach does not fit into how I do things".
--
David Kastrup
Thomas Morley writes:
> Am Do., 9. Juli 2020 um 12:30 Uhr schrieb David Kastrup :
>>
>> Thomas Morley writes:
>>
>> > Hi all,
>> >
>> > if temporary Voices occur, then the main Voice may not contain
>> > rhythmic-events for the duration o
usyGrobs context property maintained by the
Grob_pq_engraver help in any manner?
--
David Kastrup
; before including, but some placement (left/right, up/down) on the page
> would be nice.
There is \epsfile, and you can use a2ping for turning a jpeg file into
an eps file.
--
David Kastrup
ion is the same, but it is packaged
differently. You can get at the markup function part of the package by
using
test =
#(define-scheme-function (enclosure content)
(markup-function? markup?)
(list enclosure #{ \markup \box #content #}))
\markup \test \circle-markup "whatever"
Markups and their commands are an abomination.
--
David Kastrup
quot;whatever"
>
> Amazing, wonderfully elegant.
>
> Is there a way to avoid the second "\markup" and "\etc" in \markup
> \test \markup \box \etc "whatever" ?
No.
> Of course I can do
>
> circlefunc = \markup\circle\etc
> \markup \test \circlefunc "whatever"
You can? Have you tried? \circlefunc here is quite equivalent to
\circle .
--
David Kastrup
>
> BestUrs
test =
#(define-scheme-function (enclosure content)
(markup-function? markup?)
(list enclosure #{ \markup \box #content #}))
\markup \test \markup \circle \with-color #red \etc "whatever"
--
David Kastrup
ience is that \partcombine is mainly useful for piano
extracts (or to some degree partitura mashing): boiling down material
without much of a view for voice relations. For choral music (and even
polyphonic piano music) I tend to prefer fixed voice relations.
--
David Kastrup
Aaron Hill writes:
> On 2020-06-29 2:17 pm, David Kastrup wrote:
>> \voices 1,3,4,2 << \soprano \\ \alto \\ \tenor \\ \bass >>
>> which is ugly enough as it is [...]
>
> Does not seem that ugly to me, but I suppose one could encapsulate it
> via the ever-han
Pierre Perol-Schneider writes:
> Le lun. 29 juin 2020 à 22:31, David Kastrup a écrit :
>
>>
>> Just keep most stuff in the main voice, like
>>
> ...
>
>> \voices "main", 2 << { g4 f\) } \\ a,2 >>
>
> ...
> WTF???
> 'jus
d result.
>
> If it matters, I'm using version 2.20.
Just keep most stuff in the main voice, like
\fixed c' \new Voice = "main"
{
\key g \major
8.( 16)
\voiceOne
2\( 4~ |
\voices "main", 2 << { g4 f\) } \\ a,2 >>
\oneVoice
c4( d)
}
--
David Kastrup
on 4
>
> does.)
So let it expect a music expression only. 4 is a pitchless note event.
Interestingly,
zing =
#(define-void-function (x) (ly:music?)
#f)
\fixed c' { c1 \zing c4 c \zing 2 c c }
does not actually work. But that arguably is a bug that should get
fixed. If a duration is accepted as a note, it should affect the
default note duration.
--
David Kastrup
n () ()
> (let* ((tempmusic #{ {a} #})
> (els (ly:music-property tempmusic 'elements)))
> (ly:music-property (car els) 'duration)))
>
> But it does not seem to be possible to _set_ this "current duration"
> from scheme, does it?
Why would that be a good idea?
--
David Kastrup
ace, does not appear to be exposed to Scheme
> as far as I can tell.
Well, it's more a bug like anything else I guess, but
(ly:music-property #{ q #} 'duration)
at the current point of time would appear to do the trick.
--
David Kastrup
w?
>
>
> Preferred syntax: r/g or s/g
Try . You'll still need to fiddle with the chord naming function.
> Because a rest with a bass is only bass.
>
> E.g.
> \chordmode {
> a2:m r4/g r/f |
> g2 a:m |
> a:m r4/f r/g |
> g2 f |
> d:m f |
> g r4/g r8/c r/b |
> a2:m a:m |
> }
>
>
> Best regards,
> Hraban
>
>
>
--
David Kastrup
David Kastrup writes:
> Paul Scott writes:
>
>> On 6/23/20 4:01 PM, David Kastrup wrote:
>>> Paul Scott writes:
>>>
>>>> On 6/23/20 2:23 PM, David Kastrup wrote:
>>>>> Paul Scott writes:
>>>>>
>>>>>>
Paul Scott writes:
> On 6/23/20 4:01 PM, David Kastrup wrote:
>> Paul Scott writes:
>>
>>> On 6/23/20 2:23 PM, David Kastrup wrote:
>>>> Paul Scott writes:
>>>>
>>>>> The 1st argument for note is now a duration rather than
David Kastrup writes:
> Paul Scott writes:
>
>> On 6/23/20 2:23 PM, David Kastrup wrote:
>>> Paul Scott writes:
>>>
>>>> The 1st argument for note is now a duration rather than a string.
>>>>
>>>> I had some scheme def
Paul Scott writes:
> On 6/23/20 2:23 PM, David Kastrup wrote:
>> Paul Scott writes:
>>
>>> The 1st argument for note is now a duration rather than a string.
>>>
>>> I had some scheme definitions which were variations on \tempo.
>>>
>>&
I have been searching the LilyPond docs and scheme code without
> success so far.
The appendices of the Notation Reference contain lists of predicates.
--
David Kastrup
ces "first", 2
<<
{ \voiceOne c'2 g \oneVoice } \\
{ g f }
>>
c4 d e f |
\voices "first", 2
<<
{ \voiceOne c'2 g \oneVoice } \\
{ g f }
>>
}
}
\new Lyrics = "firstVoice" {
side of which you can attach ) to a
particular note. < > can even be used inside of << >> if you really
need that parallel music construct, like
< >>
By the way, that does not look like the most readable formatting.
--
David Kastrup
a-. r a-.] r a16 (as b c \bar "||"|%m4-7
>
> }
\mp has to come after notes. So has the [ you write after it. And your
last line contains a ( which is not followed by a ) possibly because you
abbreviated your example.
>
>
> \score {
> <<
>
> \new Staff { \melody }
>
> >>
> \layout { }
> \midi { }
> }
>
>
>
--
David Kastrup
elated properties to
> use "bounds" as opposed to "size".
Minimum bounds? Frankly, the description is rather useless. Without
looking up the actual code, I would have no idea what the two numbers
here are supposed to signify, respectively.
--
David Kastrup
;http://lilypond.org/doc/v2.20/Documentation/notation/displaying-rhythms#polymetric-notation>
--
David Kastrup
and the bass voice has two
bars. You don't explain how the two bars should fit the three bars.
--
David Kastrup
701 - 800 of 7439 matches
Mail list logo
