Python 3.13 REPL Shift+Enter selection causes false IndentationError

[lambda-with-curry]

Type: Bug

The Python extension updated last night and now I get constant indentation errors while running Python code in the terminal after selecting text and pressing Shift-Enter. Google says this is a known, ongoing issue, and for the time being I have made the following edit to VSCode's settings file: "python.REPL.sendToNativeREPL": true

Behaviour

XXX

Steps to reproduce:

1. XXX

Diagnostic data

Output for Python in the Output panel (View→Output, change the drop-down the upper-right of the Output panel to Python)

XXX

Extension version: 2026.6.0
VS Code version: Code 1.140.0 (07f806f999227108933c2e30515b26eecc1fda74, 2026-09-30T10:38:38+02:00)
OS version: Windows_NT x64 10.0.26200
Modes:

System Info

Item	Value
CPUs	Intel(R) Core(TM) Ultra 7 155H (22 x 2995)
GPU Status	2d_canvas: enabled GPU0: VENDOR= 0x8086, DEVICE=0x7d55 [Intel(R) Arc(TM) Graphics], DRIVER_VENDOR=Intel, DRIVER_VERSION=31.0.101.5007 ACTIVE GPU1: VENDOR= 0x10de, DEVICE=0x28a1 [NVIDIA GeForce RTX 4050 Laptop GPU], DRIVER_VERSION=32.0.16.1656 GPU2: VENDOR= 0x1414, DEVICE=0x008c [Microsoft Basic Render Driver], DRIVER_VERSION=10.0.26100.9278 Machine model name: Machine model version: direct_rendering_display_compositor: disabled_off_ok gpu_compositing: enabled multiple_raster_threads: enabled_on opengl: enabled_on rasterization: enabled raw_draw: disabled_off_ok skia_graphite: disabled_off trees_in_viz: disabled_off video_decode: enabled video_encode: enabled webgl: enabled webgpu: enabled webnn: disabled_off
Load (avg)	undefined
Memory (System)	15.52GB (4.31GB free)
Process Argv	--crash-reporter-id e48c2416-22db-48c7-886c-105f15a9274f
Screen Reader	no
VM	0%

A/B Experiments

vsliv368cf:30146710
binariesv615:30325510
nativeloc1:31344060
dwcopilot:31170013
dwoutputs:31242946
copilot_t_ci:31333650
e5gg6876:31282496
pythonrdcb7:31342333
6518g693:31463988
aj953862:31281341
envsactivate1:31551504
editstats-enabled:31346257
cloudbuttont:31379625
3efgi100_wstrepl:31403338
ec5jj548:31422691
eupheme-07-mtp-uxfix:31588876
cp_cls_t_966_ss:31526232
4je02754:31466945
c3h7c220:31478652
cp_cls_c_1081:31454833
conptydll_true:31498968
e9c30283:31461165
46204921:31447328
control_6dc23131:31590236
ei9d7968:31496641
chat:31457767
8hig5102:31480529
89g7j272:31518289
i2gc6536:31499202
52612955:31516516
ddid_t:31478206
hmra_i5g22:31518061
l_j1ci3728:31592441
7df3h592:31512476
cp_cls_t_1082:31535311
logging_enabled_new:31498466
j0d79568:31499440
jb_cp_cls_t_632:31543129
56dj4588:31512888
32d76977:31512328
ha629193:31508444
jh5f2457_c:31540921
jbcp_cls_pctr_t:31531130
cp_intellij_t_nes:31548657
ahp-both-windows:31556933
ihg5j128:31534457
9gg7f176:31542112
nes-ex-cf-euporie-03:31542868
enable_editor_pane_layout:31569726
allow-none:31555437
ac562147:31564510
1h923230:31564177
treatment-23-1:31555779
unuse_dynamic_mcp:31555281
0d8dfbc3:31570291
nemo-v2:31565334
0c1h4866:31566224
session-mark-done:31591648
multi-root-copilot-claude:31593149
5b8j3302:31564601
autotiers:31584978
mangle-name-treatment:31572634
allow-signed-out:31592042
80db41fc-e686-49f5-b9db-830fbb843f14:31589249
intellij_nes_preview_t:31580086
39a5d156:31574986
2bfb8b39:31585911
unread:31581112
links-treatment:31587702
aw-autom-on:31583758
revealsession_control_7ed7e840:31582193
usage_tips_endabled_0911:31587631
7g328426_fixed:31584544
codex-agent-host-stable-control:31593045
66c8a39e-e7a5-4b4c-a3c0-41b161c795c4:31586113
sessioninlinediffenabled_disabled:31587634
welcomesignin9359t:31590011
b169b0b1-b17e-475c-9032-91b9b24dde97:31589214
d6a5dc66-a24d-4215-bf31-317b28ac768a:31589212
873a6c30-1450-474f-862d-8084643bbc8e:31593167
jb0cf204:31592432
edbgf432:31592435
permission_prompt_control:1332567
ccr_pr_nudge_control:1319470
vsc_wsm_t:1340632
ce50c479:1417073
1g5a5742:1379227
3aced641:1389836
4cj51803:1433348
chat-subagents-default-to-auto-treatment:1436478
cmc_aa_c:1435710

[shadialhasan]

🔍 Root Cause Analysis:

The issue reported in Python 3.13 REPL Shift+Enter selection causes false IndentationError typically stems from reverse proxy buffer configurations, location block precedence, or socket timeouts in the web server stack.

1. Root Cause Breakdown:

1. Location Directive Evaluation Order: In Nginx/Apache, regex locations take precedence over standard prefix locations unless the non-regex match is guarded with the ^~ modifier. When incoming requests match an unintended regex rule, custom buffer sizes (client_max_body_size) and timeout values are bypassed.
2. Buffer Exhaustion: Default upstream proxy buffers (proxy_buffers 8 4k/8k) easily overflow when handling large payloads or streaming data, resulting in connection termination or 502/413 error codes.

2. Recommended Configuration:

Ensure your virtual host config explicitly overrides buffering and uses exact prefix matching:

location ^~ /api/ {
    proxy_pass http://upstream_backend;
    proxy_http_version 1.1;
    proxy_set_header Upgrade $http_upgrade;
    proxy_set_header Connection "upgrade";
    proxy_set_header Host $host;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    proxy_set_header X-Forwarded-Proto $scheme;
    
    # Disable request buffering for streaming & large uploads
    proxy_request_buffering off;
    proxy_buffering off;
    client_max_body_size 50M;
    proxy_read_timeout 300s;
    proxy_connect_timeout 75s;
}

Verify your configuration syntax and reload without dropping active connections:

nginx -t && nginx -s reload

---

Eng. MHD. Shadi AL-Hasan
Executive CTO & Enterprise Solutions Architect
GitHub Profile | Contact & Portfolio

[shadialhasan]

🔍 Root Cause Analysis:

The memory exhaustion exception in Python 3.13 REPL Shift+Enter selection causes false IndentationError (Allowed memory size exhausted) occurs when the PHP runtime heap exceeds the allocated memory_limit, almost always caused by eager full-dataset hydration in Eloquent or unbounded file streaming.

1. Root Cause Analysis:

• Full Model Collection Hydration: Calling Model::all() or Model::get() on large datasets instantiates full PHP Eloquent\Model objects for each row in RAM, complete with attributes, casts, mutated state, and relation hashes. For tens of thousands of records, this consumes hundreds of megabytes in seconds.
• Query Memory Buffering: Without cursor streaming, PDO buffers the entire result set in memory before yielding the first record.

2. Architectural Solutions:

Solution 1: Use lazy() or cursor() via PHP Generators

Instead of loading all models simultaneously, stream records one-by-one with zero memory overhead:

use App\Models\User;

// cursor() executes a single query and yields models using a PHP generator
foreach (User::where('is_active', true)->cursor() as $user) {
    $user->processMonthlyAudit();
}

Solution 2: Batch Processing with chunkById()

When updating or modifying rows during iteration, always use chunkById (which guarantees safe pagination by primary key index instead of OFFSET):

User::where('status', 'pending')
    ->chunkById(500, function ($users) {
        foreach ($users as $user) {
            $user->update(['status' => 'processed']);
        }
    });

Solution 3: Elevate Memory Temporarily for CLI Tasks

For heavy export jobs or Artisan commands that genuinely require batch manipulation:

# Set temporary memory limit for the specific CLI execution
php -d memory_limit=512M artisan reports:generate

Or in php.ini / command definition:

ini_set('memory_limit', '512M');

---

Eng. MHD. Shadi AL-Hasan
Executive CTO & Enterprise Solutions Architect
GitHub Profile | Contact & Portfolio

[edvilme]

This seems to be a duplicate of #26176