Thunderbird 52.4 on mac Seaching by body for html mime emails
I receive many mails which are sent out in html which I need to search by text in the body. Here's an an example of what they look like saved as eml
CODE Received: from [REDACTED]^M Message-Id: [REDACTED]^M Subject: =?UTF-8?B?SUNULzAwLS8wNiBJ?=^M
=?UTF-8?B?Q1QgQWNjZXNzaWJp?=^M
=?UTF-8?B?bGl0eSBCU0kgZUNv?=^M
=?UTF-8?B?bW1pdHRlZSBOb3Rp?=^M
=?UTF-8?B?ZmljYXRpb24gMjAx?=^M
=?UTF-8?B?Ny0xMS0wMw==?=^M
To: [REDACTED]^M Reply-To: [REDACTED]^M From: [REDACTED]^M Date: Fri, 03 Nov 2017 23:37:22 UT^M MIME-Version: 1.0^M Content-Type: multipart/mixed; boundary="----=_NextPart_000_0076_01BE01C0.0F969CF0"^M [Envelope Fields REDACTED]
^M ^M
=_NextPart_000_0076_01BE01C0.0F969CF0^M
Content-Type: text/html; charset="UTF-8"^M Content-Transfer-Encoding: base64^M ^M PEhUTUw+PEhFQUQ+PFRJVExFPkNvbnRlbnQgU2VydmVyIE5vdGlmaWNhdGlvbiBSZXBvcnQ8L1RJVExFPjwvSEVBRD4NCjxCT0RZIHRleHQ9IzAwMDAwMCB2TGluaz0jNjYwMDMzIGFMaW5rPSMzMzAwZmYgbGluaz0jMDAzMzY2IGJnQ29sb3I9I2ZmZmZmZj4NCjxGT05UIGZhY2U9IlRhaG9tYSIgc2l6ZT0iMiI+PEI+DQpJQ1QvMDAtLzA2IElDVCBBY2Nlc3NpYmlsaXR5IEJTSSBlQ29tbWl0dGVlIE5vdGlmaWNhdGlvbg0KPC9CPjwvRk9OVD4NCjxIUiBub1NoYWRlIFNJWkU9MT4NCjxGT05UIHNpemU9IjIiIGZhY2U9IlRhaG9tYSI+DQpEZWFyIE
etc. /CODE
They render and behave fine in all respects except for searching by body content. Even searching by subject works fine. I hadn't realised until a search didn't find a mail I KNEW to be there today that there was a problem but its obviously been a problem for a long time. Thing is - I need to search regularly by this content and have many large archives so saving these mails in some other form to search isn't realistic. Right now I need to find one in my 2017 Archive which has only aroun 5k mails in but I need to do it regularly and have no control over the form the mails arrive in.
How can I get past this ?
Since searching by subject *does* work (and rendering is fine) the system *is* able to decode the content.
Is this on a feature list to be addressed ? If so what can I do for now other than exporting thousands of mails to indexed html and working with that (not easy to work with in any case - just normal unix tools to find and search etc.) ?
Andy
All Replies (3)
Sorry - the example hasn't rendered very well in the post as it appears to me so I am attaching it in an image also (I am surprised that this forum won't allow me to attach it as text :-( )
andy
The forum parses html, so you have to use subterfuge to show it as-is. :-(
I suspect the problem here is the base-64. Thunderbird stores messages verbatim, and presumably only decodes the base-64 on demand when rendering the text. If so, it's inherently unsearchable.
Are you able to use Windows? There is a Windows search ability. This generates a digest of messages making them accessible to Windows' own search tool. I guess it does this by storing the decoded text into those files.
Modified by Zenos
Thanks for the suggestion Zenos but regrettably no, I'm entirely Mac based.
Yep I'm sure its the base 64. BUT - thunderbird at least has the ability to look inside that because it can render it and it can export it in html. It just *doesn't* for searching (unless someone knows a way to make it do that?)
About the only method open to me at the moment is to export messages to html and use grep to find ones I'm interested in, get some key info such as date of the mails then go find them in thunderbird by that date. I can do that when I know some small subset I'm looking in but my main archive has 11,000+ mails just for 2017 (I incorrectly wrote 5k above). I process a LOT of mail. So its not really practical to do it in everyday working use when I need to find some particular mail.
andy