Delicious Digg reddit Facebook StumbleUpon

In my previous entry I complained that opensource developers write the software that they want, which isn’t necessarily the same as the software that the users want.

I interpreted the responses to my change request as “We don’t agree with you enough to make the change”, where as what they actually meant was “We don’t agree with you strongly enough to be bothered to make the change ourselves” – the upshot of this being that when I went ahead and hacked the code myself to remove the annoying double-dash separator, the response I got was; “Could you finish the job and make this into a patch that can actually be merged back into the codebase?”. In other words, I got roped into fixing the problem properly.

In my previous entry I posted a patch that would allow you to remove the double-dash signature separator from Thunderbird if you were willing to recompile it from source yourself. I’ve now patched it properly so that there’s a new configuration option to allow anyone to change this setting at runtime – in other words I’ve done myself what I was trying to get the Thunderbird development team to do. My patch is now awaiting review and will hopefully be merged into the Thunderbird codebase in time to make the next major release (version 3).

I, as a developer, have added a feature to Thunderbird that I wanted, a feature that plenty of other people also wanted, but because no other developer had taken it upon himself to implement this feature; those who wanted it have been waiting 9 years! The point being, as a user, you only get the features you want if you’re also a developer, or if a developer happens to agree with you.

The change is actually pretty trivial – the default behaviour remains completely unchanged, and if you create a signature in Thunderbird it will still automatically add the double-dash separator before the signature. However, as a result of my patch, it will soon be possible to disable this irritating behaviour by manually setting a special configuration preference using Thunderbird’s built-in config editor. This is the preference you need to use (where ? is a number between 1 and the total number of identities you have set up) – it’s a boolean field, set it to true to disable the behaviour:

mail.identity.id?.suppress_signature_separator

And for the sake of completeness, below you’ll find the patch that I’ve submitted for inclusion in the next release of Thunderbird. Hopefully you won’t actually need to use this because it will be merged into the Thunderbird codebase.

diff -r 20c2d9e8e9b4 mailnews/base/public/nsIMsgIdentity.idl
--- a/mailnews/base/public/nsIMsgIdentity.idl	Thu Dec 31 13:43:40 2009 +0530
+++ b/mailnews/base/public/nsIMsgIdentity.idl	Sat Jan 02 20:00:55 2010 +0000
@@ -135,16 +135,21 @@ interface nsIMsgIdentity : nsISupports {
   attribute AString htmlSigText;

   /**
    * Does htmlSigText contain HTML? Use plain text if false.
    */
   attribute boolean htmlSigFormat;

   /**
+   * Suppress the double-dash signature separator
+   */
+  attribute boolean suppressSigSep;
+
+  /**
    * The encoded string representing the vcard.
    */
   attribute ACString escapedVCard;

   attribute boolean doFcc;
   /// URI for the fcc (Sent) folder
   attribute ACString fccFolder;
   attribute boolean fccReplyFollowsParent;
diff -r 20c2d9e8e9b4 mailnews/base/util/nsMsgIdentity.cpp
--- a/mailnews/base/util/nsMsgIdentity.cpp	Thu Dec 31 13:43:40 2009 +0530
+++ b/mailnews/base/util/nsMsgIdentity.cpp	Sat Jan 02 20:00:55 2010 +0000
@@ -193,16 +193,18 @@ NS_IMPL_IDPREF_BOOL(FccReplyFollowsParen
 NS_IMPL_IDPREF_STR(DraftsFolderPickerMode, "drafts_folder_picker_mode")
 NS_IMPL_IDPREF_STR(ArchivesFolderPickerMode, "archives_folder_picker_mode")
 NS_IMPL_IDPREF_STR(TmplFolderPickerMode, "tmpl_folder_picker_mode")

 NS_IMPL_IDPREF_BOOL(BccSelf, "bcc_self")
 NS_IMPL_IDPREF_BOOL(BccOthers, "bcc_other")
 NS_IMPL_IDPREF_STR (BccList, "bcc_other_list")

+NS_IMPL_IDPREF_BOOL(SuppressSigSep, "suppress_signature_separator")
+
 NS_IMETHODIMP
 nsMsgIdentity::GetDoBcc(PRBool *aValue)
 {
   nsresult rv = mPrefBranch->GetBoolPref("doBcc", aValue);
   if (NS_SUCCEEDED(rv))
     return rv;

   PRBool bccSelf = PR_FALSE;
@@ -571,16 +573,17 @@ nsMsgIdentity::Copy(nsIMsgIdentity *iden
     COPY_IDENTITY_INT_VALUE(identity,GetReplyOnTop,SetReplyOnTop)
     COPY_IDENTITY_BOOL_VALUE(identity,GetSigBottom,SetSigBottom)
     COPY_IDENTITY_BOOL_VALUE(identity,GetSigOnForward,SetSigOnForward)
     COPY_IDENTITY_BOOL_VALUE(identity,GetSigOnReply,SetSigOnReply)
     COPY_IDENTITY_INT_VALUE(identity,GetSignatureDate,SetSignatureDate)
     COPY_IDENTITY_BOOL_VALUE(identity,GetAttachVCard,SetAttachVCard)
     COPY_IDENTITY_STR_VALUE(identity,GetEscapedVCard,SetEscapedVCard)
     COPY_IDENTITY_STR_VALUE(identity,GetSmtpServerKey,SetSmtpServerKey)
+    COPY_IDENTITY_BOOL_VALUE(identity,GetSuppressSigSep,SetSuppressSigSep)
     return NS_OK;
 }

 NS_IMETHODIMP
 nsMsgIdentity::GetRequestReturnReceipt(PRBool *aVal)
 {
   NS_ENSURE_ARG_POINTER(aVal);

diff -r 20c2d9e8e9b4 mailnews/compose/src/nsMsgCompose.cpp
--- a/mailnews/compose/src/nsMsgCompose.cpp	Thu Dec 31 13:43:40 2009 +0530
+++ b/mailnews/compose/src/nsMsgCompose.cpp	Sat Jan 02 20:00:55 2010 +0000
@@ -4040,25 +4040,28 @@ nsMsgCompose::ProcessSignature(nsIMsgIde
   PRBool        attachFile = PR_FALSE;
   PRBool        useSigFile = PR_FALSE;
   PRBool        htmlSig = PR_FALSE;
   PRBool        imageSig = PR_FALSE;
   nsAutoString  sigData;
   nsAutoString sigOutput;
   PRInt32      reply_on_top = 0;
   PRBool       sig_bottom = PR_TRUE;
+  PRBool        suppressSigSep = PR_FALSE;

   nsCOMPtr sigFile;
   if (identity)
   {
     if (!CheckIncludeSignaturePrefs(identity))
       return NS_OK;

     identity->GetReplyOnTop(&reply_on_top);
     identity->GetSigBottom(&sig_bottom);
+    identity->GetSuppressSigSep(&suppressSigSep);
+
     rv = identity->GetAttachSignature(&attachFile);
     if (NS_SUCCEEDED(rv) && attachFile)
     {
       rv = identity->GetSignature(getter_AddRefs(sigFile));
       if (NS_SUCCEEDED(rv) && sigFile) {
         rv = sigFile->GetNativePath(sigNativePath);
         if (NS_SUCCEEDED(rv) && !sigNativePath.IsEmpty()) {
           PRBool exists = PR_FALSE;
@@ -4118,17 +4121,17 @@ nsMsgCompose::ProcessSignature(nsIMsgIde
   if (imageSig)
   {
     // We have an image signature. If we're using the in HTML composer, we
     // should put in the appropriate HTML for inclusion, otherwise, do nothing.
     if (m_composeHTML)
     {
       sigOutput.AppendLiteral(htmlBreak);
       sigOutput.AppendLiteral(htmlsigopen);
-      if (reply_on_top != 1 || sig_bottom || !aQuoted)
+      if (!suppressSigSep && reply_on_top != 1 || sig_bottom || !aQuoted)
         sigOutput.AppendLiteral(dashes);
       sigOutput.AppendLiteral(htmlBreak);
       sigOutput.AppendLiteral("");
@@ -4205,17 +4208,17 @@ nsMsgCompose::ProcessSignature(nsIMsgIde

     if ((reply_on_top != 1 || sig_bottom || !aQuoted) &&
         sigData.Find("\r-- \r", PR_TRUE) < 0 &&
         sigData.Find("\n-- \n", PR_TRUE) < 0 &&
         sigData.Find("\n-- \r", PR_TRUE) < 0)
     {
       nsDependentSubstring firstFourChars(sigData, 0, 4);

-      if (!(firstFourChars.EqualsLiteral("-- \n") ||
+      if (!suppressSigSep && !(firstFourChars.EqualsLiteral("-- \n") ||
             firstFourChars.EqualsLiteral("-- \r")))
       {
         sigOutput.AppendLiteral(dashes);

         if (!m_composeHTML || !htmlSig)
           sigOutput.AppendLiteral(CRLF);
         else if (m_composeHTML)
           sigOutput.AppendLiteral(htmlBreak);

This entry was posted on Saturday, January 2nd, 2010 at 10:46 pm. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.