Failed to open link. Cannot match graph identifier `logseq_notebook` to any linked graph.
Explorare opened this issue · 5 comments
[Issue]
Logseq failed to open the URI with an error message:
Failed to open link. Cannot match graph identifier
logseq_notebook
to any linked graph.
Search results are correct.
[Env]
fire_seq_search_server v0.0.5
fireSeqSearch for Firefox v0.0.6
Logseq v0.6.9
Windows 10 21H1
Hi @Explorare , thank you for reporting this issue. May I ask for a few more questions about it?
- How did you launch fire_seq_search_server? PowerShell? What's the command you were using?
- I assume you can see the correct titles on the web page. Is my guess correct?
There is a parameter --notebook_name
, I added it, but not documented it yet.
Could you please try
.\fire_seq_search_server.exe --notebook_path C:\Users\li\logseq_notebook --notebook_name LogseqNotebookName
I tried to run the server through cmd and PowerShell 7, and got the same result.
$ .\fire_seq_search_server.exe --notebook_path C:\Users\explo\Documents\Logseq --notebook_name Logseq
[INFO ] save metas
[INFO ] Preparing commit
[INFO ] Starting merge - [Seg("274d9eca"), Seg("10596cb9"), Seg("0e5df311"), Seg("41750180"), Seg("9e666f97"), Seg("619b8dc7"), Seg("e4e13444"), Seg("32388e75")]
[INFO ] Prepared commit 284
[INFO ] committing 284
[INFO ] save metas
[INFO ] Running garbage collection
[INFO ] Garbage collect
[INFO ] Server::run; addr=127.0.0.1:3030
[INFO ] listening on http://127.0.0.1:3030
[INFO ] Searching nlp transformer
[INFO ] Found doc addr DocAddress { segment_ord: 0, doc_id: 39 }, score 15.701874
[INFO ] Found doc addr DocAddress { segment_ord: 5, doc_id: 20 }, score 10.664612
[INFO ] Found doc addr DocAddress { segment_ord: 3, doc_id: 24 }, score 10.265078
[INFO ] Found doc addr DocAddress { segment_ord: 2, doc_id: 0 }, score 10.25067
[INFO ] Found doc addr DocAddress { segment_ord: 5, doc_id: 27 }, score 8.030888
[INFO ] Found doc addr DocAddress { segment_ord: 1, doc_id: 5 }, score 5.200621
[INFO ] Found doc addr DocAddress { segment_ord: 4, doc_id: 5 }, score 2.7373874
[INFO ] Found doc addr DocAddress { segment_ord: 5, doc_id: 11 }, score 1.9310763
[INFO ] Found doc addr DocAddress { segment_ord: 7, doc_id: 11 }, score 1.0827819
[INFO ] Search result ["{\"title\":[\"Transformer Neural Network\"]}","{\"title\":[\"Natural Language Processing\"]}","{\"title\":[\"Research Proposal\"]}","{\"title\":[\"@exploring transformer-based contextual knowledge graph embeddings_ how the design of the attention mask and the input structure affect learning in transformer models\"]}","{\"title\":[\"Response\"]}","{\"title\":[\"Color\"]}","{\"title\":[\"clips.What is a Knowledge Graph_\"]}","{\"title\":[\"hls__Holmström_2021_Exploring Transformer-Based Contextual Knowledge Graph Embeddings\"]}","{\"title\":[\"Knowledge Graph\"]}"]
Click the link in browser won't bring the windows of Logseq to front. And the error message showed up.
I've tried to re-index the graph but didn't help. Tried to create a new repo but didn't help.
There is a new line append to log of Logseq every time I click the link
[2022-05-21 08:21:56.076] [info] open-url {:url "logseq://graph/logseq_notebook?page=Transformer%20Neural%20Network"}
And the URI for a page that Logseq provided is
[2022-05-21 08:33:04.305] [info] open-url {:url "logseq://graph/Logseq?page=Research%20Proposal"}
So I guess the variable logseq_notebook
didn't get the right value?
My bad! Sorry for my foolish bug.
The fix is in v0.0.8. Just wait Mozilla to approve my update https://addons.mozilla.org/en-US/developers/addon/fireseqsearch/edit
Commit 62b70ef didn't fix the issue completely. The value of notebook_name
returned by fire_seq_search_server
is set to notebook_path
by mistake.
[2022-05-21 11:51:58.521] [info] open-url {:url "logseq://graph/C:%5CUsers%5Cexplo%5CDocuments%5CLogseq?page=Natural%20Language%20Processing"}
[INFO ] fire_seq_search guess the notebook name is C:\Users\explo\Documents\Logseq
I forgot to append the --notebook_name
option. With the correct value of notebook_name
been set, it works as intended. Thanks for the hotfix.