clone of my dotfiles.ssp.sh
install duckdb as preview
+350
+7
yazi/package.toml
+7
yazi/package.toml
+22
yazi/plugins/duckdb.yazi/LICENSE
+22
yazi/plugins/duckdb.yazi/LICENSE
···
1
+
# MIT License
2
+
3
+
Copyright (c) 2025 Thomas Wylie
4
+
5
+
Permission is hereby granted, free of charge, to any person obtaining a copy
6
+
of this software and associated documentation files (the "Software"), to deal
7
+
in the Software without restriction, including without limitation the rights
8
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+
copies of the Software, and to permit persons to whom the Software is
10
+
furnished to do so, subject to the following conditions:
11
+
12
+
The above copyright notice and this permission notice shall be included in all
13
+
copies or substantial portions of the Software.
14
+
15
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+
SOFTWARE.
22
+
+76
yazi/plugins/duckdb.yazi/README.md
+76
yazi/plugins/duckdb.yazi/README.md
···
1
+
# duckdb.yazi
2
+
3
+
[duckdb](https://github.com/duckdb/duckdb) now in [yazi](https://github.com/sxyazi/yazi).
4
+
5
+
<img width="1710" alt="Screenshot 2025-03-22 at 18 00 06" src="https://github.com/user-attachments/assets/db09fff9-2db1-4273-9ddf-34d0bf087967" />
6
+
7
+
## Installation
8
+
9
+
To install, use the command:
10
+
11
+
ya pack -a wylie102/duckdb
12
+
13
+
and add to your yazi.toml:
14
+
15
+
[plugin]
16
+
prepend_previewers = [
17
+
{ mime = "text/csv", run = "duckdb" },
18
+
{ name = "*.tsv", run = "duckdb" },
19
+
{ name = "*.json", run = "duckdb" },
20
+
{ name = "*.parquet", run = "duckdb" },
21
+
]
22
+
23
+
prepend_preloaders = [
24
+
{ mime = "text/csv", run = "duckdb", multi = false },
25
+
{ name = "*.tsv", run = "duckdb", multi = false },
26
+
{ name = "*.json", run = "duckdb", multi = false },
27
+
{ name = "*.parquet", run = "duckdb", multi = false },
28
+
]
29
+
30
+
### Yazi
31
+
32
+
[Installation installations](https://yazi-rs.github.io/docs/installation)
33
+
34
+
### duckdb
35
+
36
+
[Installation instructions](https://duckdb.org/docs/installation/?version=stable&environment=cli&platform=macos&download_method=direct)
37
+
38
+
## Recommended plugins
39
+
40
+
Use with a larger preview window or maximize the preview pane plugin:
41
+
<https://github.com/yazi-rs/plugins/tree/main/toggle-pane.yazi>
42
+
43
+
## What does it do?
44
+
45
+
This plugin previews your data files in yazi using DuckDB, with two available view modes:
46
+
47
+
- Standard mode (default): Displays the file as a table.
48
+
- Summarized mode: Uses DuckDB's summarize function, enhanced with custom formatting for readability.
49
+
50
+
Supported file types:
51
+
52
+
- .csv
53
+
- .json
54
+
- .parquet
55
+
- .tsv
56
+
57
+
## New Features
58
+
59
+
- Default preview mode is now "standard."
60
+
- Preview mode can be toggled within yazi:
61
+
- Press "K" at the top of the file to toggle between "standard" and "summarized."
62
+
- Preview mode is remembered per file, even after switching files or restarting yazi.
63
+
- Performance improvements through caching:
64
+
- "Standard" and "summarized" views are cached upon first load, improving scrolling performance.
65
+
66
+
## Setup and usage changes
67
+
68
+
Previously, preview mode was selected by setting an environment variable (`DUCKDB_PREVIEW_MODE`).
69
+
70
+
The new version no longer uses environment variables. Toggle preview modes directly within yazi using the keybinding described above.
71
+
72
+
Scrolling within both views (standard and summarized) is handled by pressing J (down) and K (up). Performance is significantly better due to caching.
73
+
74
+
## Preview
75
+
76
+
<img width="1710" alt="Screenshot 2025-03-22 at 17 59 21" src="https://github.com/user-attachments/assets/ac006667-4281-4e0a-87a4-bfaeefc6f20b" />
+231
yazi/plugins/duckdb.yazi/main.lua
+231
yazi/plugins/duckdb.yazi/main.lua
···
1
+
-- This function generates the SQL query based on the preview mode.
2
+
local function generate_sql(job, mode)
3
+
if mode == "standard" then
4
+
return string.format("SELECT * FROM '%s' LIMIT 500", tostring(job.file.url))
5
+
else
6
+
return string.format(
7
+
[[SELECT
8
+
column_name AS column,
9
+
column_type AS type,
10
+
count,
11
+
approx_unique AS unique,
12
+
null_percentage AS null,
13
+
LEFT(min, 10) AS min,
14
+
LEFT(max, 10) AS max,
15
+
CASE
16
+
WHEN column_type IN ('TIMESTAMP', 'DATE') THEN '-'
17
+
WHEN avg IS NULL THEN 'NULL'
18
+
WHEN TRY_CAST(avg AS DOUBLE) IS NULL THEN avg
19
+
WHEN CAST(avg AS DOUBLE) < 100000 THEN CAST(ROUND(CAST(avg AS DOUBLE), 2) AS VARCHAR)
20
+
WHEN CAST(avg AS DOUBLE) < 1000000 THEN CAST(ROUND(CAST(avg AS DOUBLE) / 1000, 1) AS VARCHAR) || 'k'
21
+
WHEN CAST(avg AS DOUBLE) < 1000000000 THEN CAST(ROUND(CAST(avg AS DOUBLE) / 1000000, 2) AS VARCHAR) || 'm'
22
+
WHEN CAST(avg AS DOUBLE) < 1000000000000 THEN CAST(ROUND(CAST(avg AS DOUBLE) / 1000000000, 2) AS VARCHAR) || 'b'
23
+
ELSE '∞'
24
+
END AS avg,
25
+
CASE
26
+
WHEN column_type IN ('TIMESTAMP', 'DATE') THEN '-'
27
+
WHEN std IS NULL THEN 'NULL'
28
+
WHEN TRY_CAST(std AS DOUBLE) IS NULL THEN std
29
+
WHEN CAST(std AS DOUBLE) < 100000 THEN CAST(ROUND(CAST(std AS DOUBLE), 2) AS VARCHAR)
30
+
WHEN CAST(std AS DOUBLE) < 1000000 THEN CAST(ROUND(CAST(std AS DOUBLE) / 1000, 1) AS VARCHAR) || 'k'
31
+
WHEN CAST(std AS DOUBLE) < 1000000000 THEN CAST(ROUND(CAST(std AS DOUBLE) / 1000000, 2) AS VARCHAR) || 'm'
32
+
WHEN CAST(std AS DOUBLE) < 1000000000000 THEN CAST(ROUND(CAST(std AS DOUBLE) / 1000000000, 2) AS VARCHAR) || 'b'
33
+
ELSE '∞'
34
+
END AS std,
35
+
CASE
36
+
WHEN column_type IN ('TIMESTAMP', 'DATE') THEN '-'
37
+
WHEN q25 IS NULL THEN 'NULL'
38
+
WHEN TRY_CAST(q25 AS DOUBLE) IS NULL THEN q25
39
+
WHEN CAST(q25 AS DOUBLE) < 100000 THEN CAST(ROUND(CAST(q25 AS DOUBLE), 2) AS VARCHAR)
40
+
WHEN CAST(q25 AS DOUBLE) < 1000000 THEN CAST(ROUND(CAST(q25 AS DOUBLE) / 1000, 1) AS VARCHAR) || 'k'
41
+
WHEN CAST(q25 AS DOUBLE) < 1000000000 THEN CAST(ROUND(CAST(q25 AS DOUBLE) / 1000000, 2) AS VARCHAR) || 'm'
42
+
WHEN CAST(q25 AS DOUBLE) < 1000000000000 THEN CAST(ROUND(CAST(q25 AS DOUBLE) / 1000000000, 2) AS VARCHAR) || 'b'
43
+
ELSE '∞'
44
+
END AS q25,
45
+
CASE
46
+
WHEN column_type IN ('TIMESTAMP', 'DATE') THEN '-'
47
+
WHEN q50 IS NULL THEN 'NULL'
48
+
WHEN TRY_CAST(q50 AS DOUBLE) IS NULL THEN q50
49
+
WHEN CAST(q50 AS DOUBLE) < 100000 THEN CAST(ROUND(CAST(q50 AS DOUBLE), 2) AS VARCHAR)
50
+
WHEN CAST(q50 AS DOUBLE) < 1000000 THEN CAST(ROUND(CAST(q50 AS DOUBLE) / 1000, 1) AS VARCHAR) || 'k'
51
+
WHEN CAST(q50 AS DOUBLE) < 1000000000 THEN CAST(ROUND(CAST(q50 AS DOUBLE) / 1000000, 2) AS VARCHAR) || 'm'
52
+
WHEN CAST(q50 AS DOUBLE) < 1000000000000 THEN CAST(ROUND(CAST(q50 AS DOUBLE) / 1000000000, 2) AS VARCHAR) || 'b'
53
+
ELSE '∞'
54
+
END AS q50,
55
+
CASE
56
+
WHEN column_type IN ('TIMESTAMP', 'DATE') THEN '-'
57
+
WHEN q75 IS NULL THEN 'NULL'
58
+
WHEN TRY_CAST(q75 AS DOUBLE) IS NULL THEN q75
59
+
WHEN CAST(q75 AS DOUBLE) < 100000 THEN CAST(ROUND(CAST(q75 AS DOUBLE), 2) AS VARCHAR)
60
+
WHEN CAST(q75 AS DOUBLE) < 1000000 THEN CAST(ROUND(CAST(q75 AS DOUBLE) / 1000, 1) AS VARCHAR) || 'k'
61
+
WHEN CAST(q75 AS DOUBLE) < 1000000000 THEN CAST(ROUND(CAST(q75 AS DOUBLE) / 1000000, 2) AS VARCHAR) || 'm'
62
+
WHEN CAST(q75 AS DOUBLE) < 1000000000000 THEN CAST(ROUND(CAST(q75 AS DOUBLE) / 1000000000, 2) AS VARCHAR) || 'b'
63
+
ELSE '∞'
64
+
END AS q75
65
+
FROM (summarize FROM '%s')]],
66
+
tostring(job.file.url)
67
+
)
68
+
end
69
+
end
70
+
71
+
local function get_cache_path(job, type)
72
+
local skip = job.skip
73
+
job.skip = 0
74
+
local base = ya.file_cache(job)
75
+
job.skip = skip
76
+
if not base then
77
+
return nil
78
+
end
79
+
local suffix = ({ standard = "_standard.db", summarized = "_summarized.db", mode = "_mode.db" })[type or "standard"]
80
+
return Url(tostring(base) .. suffix)
81
+
end
82
+
83
+
local function run_query(job, query, target)
84
+
local args = {}
85
+
if target ~= job.file.url then
86
+
table.insert(args, tostring(target))
87
+
end
88
+
table.insert(args, "-c")
89
+
table.insert(args, query)
90
+
local child = Command("duckdb"):args(args):stdout(Command.PIPED):stderr(Command.PIPED):spawn()
91
+
if not child then
92
+
return nil
93
+
end
94
+
local output, err = child:wait_with_output()
95
+
if err then
96
+
return nil
97
+
end
98
+
if not output.status.success then
99
+
ya.err("DuckDB exited with error: " .. output.stderr)
100
+
return nil
101
+
end
102
+
return output
103
+
end
104
+
105
+
local function create_cache(job, mode, path)
106
+
local filename = job.file.url:name() or "unknown"
107
+
if fs.cha(path) then
108
+
return true
109
+
end
110
+
local sql = (mode == "mode") and "CREATE TABLE My_table AS SELECT 'standard' AS Preview_mode;"
111
+
or string.format("CREATE TABLE My_table AS (%s);", generate_sql(job, mode))
112
+
local out = run_query(job, sql, path, mode == "mode" and "mode" or nil)
113
+
if not out then
114
+
ya.err("Preload - Failed to generate " .. mode .. " cache for file: " .. tostring(filename) .. ".")
115
+
return false
116
+
end
117
+
return true
118
+
end
119
+
120
+
local function get_preview_mode(job)
121
+
local mode = "standard"
122
+
local mode_cache = get_cache_path(job, "mode")
123
+
if not mode_cache then
124
+
return mode
125
+
end
126
+
if not fs.cha(mode_cache) then
127
+
create_cache(job, "mode", mode_cache)
128
+
end
129
+
local result = run_query(job, "SELECT Preview_mode FROM My_table LIMIT 1;", mode_cache, "mode")
130
+
if result and result.stdout and result.stdout ~= "" then
131
+
local value = result.stdout:lower()
132
+
if value:match("summarized") then
133
+
mode = "summarized"
134
+
end
135
+
end
136
+
return mode
137
+
end
138
+
139
+
local function generate_query(target, job, limit, offset)
140
+
local mode = get_preview_mode(job)
141
+
if target == job.file.url then
142
+
if mode == "standard" then
143
+
return string.format("SELECT * FROM '%s' LIMIT %d OFFSET %d;", tostring(target), limit, offset)
144
+
else
145
+
local query = generate_sql(job, mode)
146
+
return string.format("WITH query AS (%s) SELECT * FROM query LIMIT %d OFFSET %d;", query, limit, offset)
147
+
end
148
+
else
149
+
return string.format("SELECT * FROM My_table LIMIT %d OFFSET %d;", limit, offset)
150
+
end
151
+
end
152
+
153
+
local function set_preview_mode(job, mode)
154
+
local mode_cache = get_cache_path(job, "mode")
155
+
if not mode_cache then
156
+
return false
157
+
end
158
+
run_query(job, "DELETE FROM My_table;", mode_cache, "mode")
159
+
local sql = string.format("INSERT INTO My_table VALUES ('%s');", mode)
160
+
local result = run_query(job, sql, mode_cache, "mode")
161
+
if not result then
162
+
ya.err("SetPreviewMode - Failed to update preview mode.")
163
+
return false
164
+
end
165
+
return true
166
+
end
167
+
168
+
local M = {}
169
+
170
+
function M:preload(job)
171
+
local cache_standard = get_cache_path(job, "standard")
172
+
local cache_summarized = get_cache_path(job, "summarized")
173
+
if not cache_standard or not cache_summarized then
174
+
return false
175
+
end
176
+
if fs.cha(cache_standard) and fs.cha(cache_summarized) then
177
+
return true
178
+
end
179
+
local success = true
180
+
success = create_cache(job, "standard", cache_standard) and success
181
+
success = create_cache(job, "summarized", cache_summarized) and success
182
+
return success
183
+
end
184
+
185
+
function M:peek(job)
186
+
local raw_skip = job.skip or 0
187
+
local skip = math.max(0, raw_skip - 50)
188
+
if raw_skip > 0 and raw_skip < 50 then
189
+
local current_mode = get_preview_mode(job)
190
+
local new_mode = current_mode == "standard" and "summarized" or "standard"
191
+
set_preview_mode(job, new_mode)
192
+
skip = 0
193
+
end
194
+
job.skip = skip
195
+
local mode = get_preview_mode(job)
196
+
local cache = get_cache_path(job, mode)
197
+
local file_url = job.file.url
198
+
local target = cache
199
+
local limit = job.area.h - 7
200
+
local offset = skip
201
+
if not cache or not fs.cha(cache) then
202
+
target = file_url
203
+
end
204
+
local query = generate_query(target, job, limit, offset)
205
+
local output = run_query(job, query, target)
206
+
if not output or output.stdout == "" then
207
+
if target ~= file_url then
208
+
target = file_url
209
+
query = generate_query(target, job, limit, offset)
210
+
output = run_query(job, query, target)
211
+
if not output or output.stdout == "" then
212
+
return require("code"):peek(job)
213
+
end
214
+
else
215
+
return require("code"):peek(job)
216
+
end
217
+
end
218
+
ya.preview_widgets(job, { ui.Text.parse(output.stdout):area(job.area) })
219
+
end
220
+
221
+
function M:seek(job)
222
+
local OFFSET_BASE = 50
223
+
local encoded_current_skip = cx.active.preview.skip or 0
224
+
local current_skip = math.max(0, encoded_current_skip - OFFSET_BASE)
225
+
local units = job.units or 0
226
+
local new_skip = current_skip + units
227
+
local encoded_skip = new_skip + OFFSET_BASE
228
+
ya.manager_emit("peek", { encoded_skip, only_if = job.file.url })
229
+
end
230
+
231
+
return M
+14
yazi/yazi.toml
+14
yazi/yazi.toml
···
1
+
[plugin]
2
+
prepend_previewers = [
3
+
{ mime = "text/csv", run = "duckdb" },
4
+
{ name = ".tsv", run = "duckdb" },
5
+
{ name = ".json", run = "duckdb" },
6
+
{ name = "*.parquet", run = "duckdb" },
7
+
]
8
+
9
+
prepend_preloaders = [
10
+
{ mime = "text/csv", run = "duckdb", multi = false },
11
+
{ name = ".tsv", run = "duckdb", multi = false },
12
+
{ name = ".json", run = "duckdb", multi = false },
13
+
{ name = "*.parquet", run = "duckdb", multi = false },
14
+
]