Dear Squeakers,
I tried to parse an HTML file like this in a trunk image and ran into a MNU: FileStream fileNamed: ’some.html’ do: [:stream | HtmlParser parse: stream]
In HtmlText>>#initialize the message #replaceHtmlCharRefs is sent. I suppose this method was once the image. Otherwise HtmlParser would never have worked. How can I find out, when it got lost? How would you do it?
Cheers, Bernhard
'From Squeak3.4 of 1 March 2003 [latest update: #5170] on 29 March 2003 at 6:47:54 pm' "Change Set: RemoveScamper Date: 29 March 2003 Author: Adam Spitz
Removes Scamper from the image (assuming all references to it have already been removed)."
StringremoveSelector: #replaceHtmlCharRefs. Smalltalk organization removeCategoriesMatching: 'Network-HTML*'. Smalltalk organization removeCategoriesMatching: 'Network-Web Browser'.
Utilities informUser: 'Removing Scamper thumbnails from Tools flap and PartsBin. Please wait...' during: [ PartsBin clearThumbnailCache. PartsBin cacheAllThumbnails. Flaps replaceToolsFlap.
On 10/22/17 1:05 PM, Bernhard Pieber wrote:
Dear Squeakers,
I tried to parse an HTML file like this in a trunk image and ran into a MNU: FileStream fileNamed: ’some.html’ do: [:stream | HtmlParser parse: stream]
In HtmlText>>#initialize the message #replaceHtmlCharRefs is sent. I suppose this method was once the image. Otherwise HtmlParser would never have worked. How can I find out, when it got lost? How would you do it?
Cheers, Bernhard
If you just want to replace it yourself, try this:
'From Squeak3.4alpha of ''11 November 2002'' [latest update: #5109] on 16 November 2002 at 8:06:43 pm' "Change Set: ISO8859 Date: 15 November 2002 Author: Boris Gaertner
Jean-Marie Zajac pointed out that accented characters in ISO-8859-1 encoding are not displayed as expected. Scamper is not encoding-aware, but it translates ISO-8859-1 to the encoding that is used in Squeak. Unfortunately, due to a subtle bug the translation is done twice: first, the entire source is translated, later parsed entities are translated again. This change set drops the translation of parsed entites. To make it work, it adds the translation of character entity references (characters that are written in the form &#<integer>; or in the form &<character name>; see sections 5.3.1 and 5.3.2 of the HTML 4.0 specification.)
Jean-Marie tested a first version and found a new bug, later he tested a second version that is seemingly ok. With his test he helped me to understand where the real problem was burried. Thanks a lot!
" HtmlTextmethodsFor: 'private-initialization' stamp: 'BG 11/15/2002 21:40' initialize:source0 super initialize: source0. self text: source0replaceHtmlCharRefs. StringmethodsFor: 'internet' stamp: 'BG 11/15/2002 21:18' replaceHtmlCharRefs
| pos ampIndex scIndex special specialValue outString outPos newOutPos |
outString ← String new: self size. outPos ← 0.
pos ← 1.
[ pos <= self size ] whileTrue: [ "read up to the next ampersand" ampIndex ← self indexOf: $& startingAt: pos ifAbsent: [0].
ampIndex = 0 ifTrue: [ pos = 1 ifTrue: [ ↑self ] ifFalse: [ ampIndex ← self size+1 ] ].
newOutPos ← outPos + ampIndex - pos. outString replaceFrom: outPos + 1 to: newOutPos with: self startingAt: pos. outPos ← newOutPos. pos ← ampIndex.
ampIndex <= self size ifTrue: [ "find the $;" scIndex ← self indexOf: $; startingAt: ampIndex ifAbsent: [ self size + 1 ].
special ← self copyFrom: ampIndex+1 to: scIndex-1. specialValue ← HtmlEntity valueOfHtmlEntity: special.
specialValue ifNil: [ "not a recognized entity. wite it back" scIndex > self size ifTrue: [ scIndex ← self size ].
newOutPos ← outPos + scIndex - ampIndex + 1. outString replaceFrom: outPos+1 to: newOutPos with: self startingAt: ampIndex. outPos ← newOutPos.] ifNotNil: [ outPos ← outPos + 1. outString at: outPos put: specialValue isoToSqueak.].
pos ← scIndex + 1. ]. ].
↑outString copyFrom: 1 to: outPos
On 10/22/17 1:05 PM, Bernhard Pieber wrote:
Dear Squeakers,
I tried to parse an HTML file like this in a trunk image and ran into a MNU: FileStream fileNamed: ’some.html’ do: [:stream | HtmlParser parse: stream]
In HtmlText>>#initialize the message #replaceHtmlCharRefs is sent. I suppose this method was once the image. Otherwise HtmlParser would never have worked. How can I find out, when it got lost? How would you do it?
Cheers, Bernhard
Hi Bob,
Thanks for your answer. It helped me find the method in a 3.4 image. How did you find the ChangeSets?
Cheers, Bernhard
Am 22.10.2017 um 19:46 schrieb Bob Arning arning315@comcast.net:
If you just want to replace it yourself, try this:
'From Squeak3.4alpha of ''11 November 2002'' [latest update: #5109] on 16 November 2002 at 8:06:43 pm' "Change Set: ISO8859 Date: 15 November 2002 Author: Boris Gaertner
Jean-Marie Zajac pointed out that accented characters in ISO-8859-1 encoding are not displayed as expected. Scamper is not encoding-aware, but it translates ISO-8859-1 to the encoding that is used in Squeak. Unfortunately, due to a subtle bug the translation is done twice: first, the entire source is translated, later parsed entities are translated again. This change set drops the translation of parsed entites. To make it work, it adds the translation of character entity references (characters that are written in the form &#<integer>; or in the form &<character name>; see sections 5.3.1 and 5.3.2 of the HTML 4.0 specification.)
Jean-Marie tested a first version and found a new bug, later he tested a second version that is seemingly ok. With his test he helped me to understand where the real problem was burried. Thanks a lot!
" HtmlText methodsFor: 'private-initialization' stamp: 'BG 11/15/2002 21:40' initialize: source0 super initialize: source0. self text: source0 replaceHtmlCharRefs. String methodsFor: 'internet' stamp: 'BG 11/15/2002 21:18' replaceHtmlCharRefs
| pos ampIndex scIndex special specialValue outString outPos newOutPos |
outString ← String new: self size. outPos ← 0.
pos ← 1.
[ pos <= self size ] whileTrue: [ "read up to the next ampersand" ampIndex ← self indexOf: $& startingAt: pos ifAbsent: [0].
ampIndex = 0 ifTrue: [ pos = 1 ifTrue: [ ↑self ] ifFalse: [ ampIndex ← self size+1 ] ].
newOutPos ← outPos + ampIndex - pos. outString replaceFrom: outPos + 1 to: newOutPos with: self startingAt: pos. outPos ← newOutPos. pos ← ampIndex.
ampIndex <= self size ifTrue: [ "find the $;" scIndex ← self indexOf: $; startingAt: ampIndex ifAbsent: [ self size + 1 ].
special ← self copyFrom: ampIndex+1 to: scIndex-1. specialValue ← HtmlEntity valueOfHtmlEntity: special.
specialValue ifNil: [ "not a recognized entity. wite it back" scIndex > self size ifTrue: [ scIndex ← self size ].
newOutPos ← outPos + scIndex - ampIndex + 1. outString replaceFrom: outPos+1 to: newOutPos with: self startingAt: ampIndex. outPos ← newOutPos.] ifNotNil: [ outPos ← outPos + 1. outString at: outPos put: specialValue isoToSqueak.].
pos ← scIndex + 1. ]. ].
↑outString copyFrom: 1 to: outPos
On 10/22/17 1:05 PM, Bernhard Pieber wrote:
Dear Squeakers,
I tried to parse an HTML file like this in a trunk image and ran into a MNU: FileStream fileNamed: ’some.html’ do: [:stream | HtmlParser parse: stream]
In HtmlText>>#initialize the message #replaceHtmlCharRefs is sent. I suppose this method was once the image. Otherwise HtmlParser would never have worked. How can I find out, when it got lost? How would you do it?
Cheers, Bernhard
Andreas Raab assembled the historical update stream (change sets) and saved it on the server at http://files.squeak.org/history/
There is also the http://files.squeak.org/updates/ folder, although I think that the ./history folder is the most complete collection available.
Later updates were done with the Monticello update stream that is currently in use.
Dave
On Sun, Oct 22, 2017 at 09:42:14PM +0200, Bernhard Pieber wrote:
Hi Bob,
Thanks for your answer. It helped me find the method in a 3.4 image. How did you find the ChangeSets?
Cheers, Bernhard
Am 22.10.2017 um 19:46 schrieb Bob Arning arning315@comcast.net:
If you just want to replace it yourself, try this:
'From Squeak3.4alpha of ''11 November 2002'' [latest update: #5109] on 16 November 2002 at 8:06:43 pm' "Change Set: ISO8859 Date: 15 November 2002 Author: Boris Gaertner
Jean-Marie Zajac pointed out that accented characters in ISO-8859-1 encoding are not displayed as expected. Scamper is not encoding-aware, but it translates ISO-8859-1 to the encoding that is used in Squeak. Unfortunately, due to a subtle bug the translation is done twice: first, the entire source is translated, later parsed entities are translated again. This change set drops the translation of parsed entites. To make it work, it adds the translation of character entity references (characters that are written in the form &#<integer>; or in the form &<character name>; see sections 5.3.1 and 5.3.2 of the HTML 4.0 specification.)
Jean-Marie tested a first version and found a new bug, later he tested a second version that is seemingly ok. With his test he helped me to understand where the real problem was burried. Thanks a lot!
" HtmlText methodsFor: 'private-initialization' stamp: 'BG 11/15/2002 21:40' initialize: source0 super initialize: source0. self text: source0 replaceHtmlCharRefs. String methodsFor: 'internet' stamp: 'BG 11/15/2002 21:18' replaceHtmlCharRefs
| pos ampIndex scIndex special specialValue outString outPos newOutPos |
outString ??? String new: self size. outPos ??? 0.
pos ??? 1.
[ pos <= self size ] whileTrue: [ "read up to the next ampersand" ampIndex ??? self indexOf: $& startingAt: pos ifAbsent: [0].
ampIndex = 0 ifTrue: [ pos = 1 ifTrue: [ ???self ] ifFalse: [ ampIndex ??? self size+1 ] ].
newOutPos ??? outPos + ampIndex - pos. outString replaceFrom: outPos + 1 to: newOutPos with: self startingAt: pos. outPos ??? newOutPos. pos ??? ampIndex.
ampIndex <= self size ifTrue: [ "find the $;" scIndex ??? self indexOf: $; startingAt: ampIndex ifAbsent: [ self size + 1 ].
special ??? self copyFrom: ampIndex+1 to: scIndex-1. specialValue ??? HtmlEntity valueOfHtmlEntity: special.
specialValue ifNil: [ "not a recognized entity. wite it back" scIndex > self size ifTrue: [ scIndex ??? self size ].
newOutPos ??? outPos + scIndex - ampIndex + 1. outString replaceFrom: outPos+1 to: newOutPos with: self startingAt: ampIndex. outPos ??? newOutPos.] ifNotNil: [ outPos ??? outPos + 1. outString at: outPos put: specialValue isoToSqueak.].
pos ??? scIndex + 1. ]. ].
???outString copyFrom: 1 to: outPos
On 10/22/17 1:05 PM, Bernhard Pieber wrote:
Dear Squeakers,
I tried to parse an HTML file like this in a trunk image and ran into a MNU: FileStream fileNamed: ???some.html??? do: [:stream | HtmlParser parse: stream]
In HtmlText>>#initialize the message #replaceHtmlCharRefs is sent. I suppose this method was once the image. Otherwise HtmlParser would never have worked. How can I find out, when it got lost? How would you do it?
Cheers, Bernhard
That is a great resource! I downloaded all zip files and I could easily find a ChangeSet with the missing method. Good to know for the future.
Cheers, Bernhard
Am 22.10.2017 um 22:01 schrieb David T. Lewis lewis@mail.msen.com:
Andreas Raab assembled the historical update stream (change sets) and saved it on the server at http://files.squeak.org/history/
There is also the http://files.squeak.org/updates/ folder, although I think that the ./history folder is the most complete collection available.
Later updates were done with the Monticello update stream that is currently in use.
Dave
On Sun, Oct 22, 2017 at 09:42:14PM +0200, Bernhard Pieber wrote:
Hi Bob,
Thanks for your answer. It helped me find the method in a 3.4 image. How did you find the ChangeSets?
Cheers, Bernhard
Am 22.10.2017 um 19:46 schrieb Bob Arning arning315@comcast.net:
If you just want to replace it yourself, try this:
'From Squeak3.4alpha of ''11 November 2002'' [latest update: #5109] on 16 November 2002 at 8:06:43 pm' "Change Set: ISO8859 Date: 15 November 2002 Author: Boris Gaertner
Jean-Marie Zajac pointed out that accented characters in ISO-8859-1 encoding are not displayed as expected. Scamper is not encoding-aware, but it translates ISO-8859-1 to the encoding that is used in Squeak. Unfortunately, due to a subtle bug the translation is done twice: first, the entire source is translated, later parsed entities are translated again. This change set drops the translation of parsed entites. To make it work, it adds the translation of character entity references (characters that are written in the form &#<integer>; or in the form &<character name>; see sections 5.3.1 and 5.3.2 of the HTML 4.0 specification.)
Jean-Marie tested a first version and found a new bug, later he tested a second version that is seemingly ok. With his test he helped me to understand where the real problem was burried. Thanks a lot!
" HtmlText methodsFor: 'private-initialization' stamp: 'BG 11/15/2002 21:40' initialize: source0 super initialize: source0. self text: source0 replaceHtmlCharRefs. String methodsFor: 'internet' stamp: 'BG 11/15/2002 21:18' replaceHtmlCharRefs
| pos ampIndex scIndex special specialValue outString outPos newOutPos |
outString ??? String new: self size. outPos ??? 0.
pos ??? 1.
[ pos <= self size ] whileTrue: [ "read up to the next ampersand" ampIndex ??? self indexOf: $& startingAt: pos ifAbsent: [0].
ampIndex = 0 ifTrue: [ pos = 1 ifTrue: [ ???self ] ifFalse: [ ampIndex ??? self size+1 ] ].
newOutPos ??? outPos + ampIndex - pos. outString replaceFrom: outPos + 1 to: newOutPos with: self startingAt: pos. outPos ??? newOutPos. pos ??? ampIndex.
ampIndex <= self size ifTrue: [ "find the $;" scIndex ??? self indexOf: $; startingAt: ampIndex ifAbsent: [ self size + 1 ].
special ??? self copyFrom: ampIndex+1 to: scIndex-1. specialValue ??? HtmlEntity valueOfHtmlEntity: special.
specialValue ifNil: [ "not a recognized entity. wite it back" scIndex > self size ifTrue: [ scIndex ??? self size ].
newOutPos ??? outPos + scIndex - ampIndex + 1. outString replaceFrom: outPos+1 to: newOutPos with: self startingAt: ampIndex. outPos ??? newOutPos.] ifNotNil: [ outPos ??? outPos + 1. outString at: outPos put: specialValue isoToSqueak.].
pos ??? scIndex + 1. ]. ].
???outString copyFrom: 1 to: outPos
On 10/22/17 1:05 PM, Bernhard Pieber wrote:
Dear Squeakers,
I tried to parse an HTML file like this in a trunk image and ran into a MNU: FileStream fileNamed: ???some.html??? do: [:stream | HtmlParser parse: stream]
In HtmlText>>#initialize the message #replaceHtmlCharRefs is sent. I suppose this method was once the image. Otherwise HtmlParser would never have worked. How can I find out, when it got lost? How would you do it?
Cheers, Bernhard
Fascinating! Is this something you could maybe share? Or have you already? A quick mailing list and Google search turned up nothing.
Cheers, Bernhard
Am 22.10.2017 um 22:10 schrieb Bob Arning arning315@comcast.net:
I wrote a change browser for as much of squeak history as I could find a few years ago.
On 10/22/17 3:42 PM, Bernhard Pieber wrote:
Thanks for your answer. It helped me find the method in a 3.4 image. How did you find the ChangeSets?
It was available for a while in 2013, but interest waned. You can search messages to the squeak list in 2013 with 'changes', 'evolution' or 'archaeology' in the title.
A (somewhat dated) code snapshot is at https://www.dropbox.com/l/scl/AAD3x0OLu3lxYlc0IhiKNXKQzRLCrMZvqtA
On 10/22/17 4:16 PM, Bernhard Pieber wrote:
Fascinating! Is this something you could maybe share? Or have you already? A quick mailing list and Google search turned up nothing.
Cheers, Bernhard
Am 22.10.2017 um 22:10 schrieb Bob Arning arning315@comcast.net:
I wrote a change browser for as much of squeak history as I could find a few years ago.
On 10/22/17 3:42 PM, Bernhard Pieber wrote:
Thanks for your answer. It helped me find the method in a 3.4 image. How did you find the ChangeSets?
squeak-dev@lists.squeakfoundation.org