`jcode run` prints streamed reasoning one chunk per line
Maintainers usually reply within 2 days
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 82/100
Research direction
Start in crates/jcode-app-core/src/agent/turn_loops.rs at run_turn and follow the StreamEvent::ThinkingDelta handling. Compare it with the text delta output path, which the issue says already wraps correctly. Update the reasoning output so one block uses a single prefix and continuous text; done when streamed reasoning wraps at the terminal width in plain jcode run and the line-based REPL.
Written by the indexing model from the issue text.
Description
Curious - what was the idea behind bare jcode run anyway? It could maybe use some love to become a more parseable plain-text alternative to running a full TUI, but I'm not sure where to start - right now it looks more like a debug output of a run, rather than something an output of which can be easily read or shared.
What happens
In plain jcode run output (and the line-based REPL), streamed reasoning is
printed one provider delta per line, each with its own 💭 prefix. Deltas are
usually only a few words, so reasoning shows up as a narrow column no matter
how wide the terminal is. Same behavior inside and outside tmux.
💭 I'm starting the
💭 investigation now
💭 , ke
💭 eping to read-only work
Expected
Reasoning for one block is printed as continuous text after a single prefix,
and the terminal wraps it at its real width.
Cause
run_turn in crates/jcode-app-core/src/agent/turn_loops.rs handles
StreamEvent::ThinkingDelta with terminal_println!("💭 {}", thinking_text),
so every delta gets a prefix and a newline. Text deltas already use
terminal_print! and wrap correctly.
- Dominant language
- Rust
- Stars
- 19.9k
- Forks
- 2.3k
- Avg merge
- 21h 27m
- Merged PRs (30d)
- 7
Getting set up
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from 1jehuang/jcode
-
Difficulty 1/5 Under an hour Newbie friendliness 85/100
Maintainers usually reply within 2 days
-
ACP titles are non descriptive, even though they could bePossibly taken @Aafimalek claimed this 5 days ago. Openautonomous: likely enhancement type: feature
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
1jehuang/jcode#2093 · 1 comment ·
Maintainers usually reply within 2 days
-
autonomous: no triage: needs-decision
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
1jehuang/jcode#2079 · 1 comment ·
Maintainers usually reply within 2 days
-
autonomous: no triage: needs-decision
Difficulty 2/5 1-3 hours Newbie friendliness 75/100
1jehuang/jcode#2070 · 1 comment ·
Maintainers usually reply within 2 days
-
bug triage: fixed-pending-release
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
1jehuang/jcode#2019 · 3 comments ·
Maintainers usually reply within 2 days
Similar issues
-
scheduled-for/27.3.0
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
stackabletech/operator-rs#1331 ·
Maintainers usually reply within 1 day
-
enhancement
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
zellij-org/zellij#5929 ·
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
swc-project/swc-next-issues#396 ·
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 83/100
nautechsystems/nautilus_trader#5685 ·
Maintainers usually reply within 1 day
-
bug db/postgres user-priority/P1
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
Maintainers usually reply within 1 day