From d104d5d936a7854e3ba60d6963213f0b41e76e54 Mon Sep 17 00:00:00 2001
From: sneak <sneak@sneak.berlin>
Date: Fri, 16 Oct 2026 00:45:38 +0000
Subject: [PATCH] Escape state_dir in the db_url derived from it (closes #290)

The derived db_url put state_dir into a file: URL as it was, so SQLite
decoded a %HH in the name and ended the path at a ? or #, and the
journal mode parameter was lost. It is now built with net/url, which
percent-encodes the directory name; SQLite decodes it back and opens
state.sqlite3 in that directory. OmitHost keeps the usual form
file:/var/lib/pixa/state.sqlite3 and a relative state_dir a path.

Model: opus-5-5
---
 README.md                 | 12 +++++++-----
 TODO.md                   |  6 ++++++
 internal/config/config.go | 21 +++++++++++++++++++--
 3 files changed, 32 insertions(+), 7 deletions(-)

diff --git a/README.md b/README.md
index d9b3c86..17ceda1 100644
--- a/README.md
+++ b/README.md
@@ -725,11 +725,13 @@ Key settings in more detail:
 - `signing_key` — HMAC secret for URL signatures
 - `db_url` — the SQLite database to open; omitted, it is
   `file:<state_dir>/state.sqlite3?_pragma=journal_mode(WAL)`, which keeps the
-  database in WAL mode. pixa opens one connection to it, so its own reads and
-  writes run one at a time. pixa adds `_pragma=busy_timeout(5000)` to any
-  `db_url`, so a write that finds another program writing to the file waits up
-  to five seconds for it instead of failing. WAL mode comes only from the URL:
-  keep `_pragma=journal_mode(WAL)` in one you set
+  database in WAL mode. `state_dir` is percent-encoded in it, so a `%`, `?` or
+  `#` in the name is written as `%25`, `%3F` or `%23`. pixa opens one connection
+  to it, so its own reads and writes run one at a time. pixa adds
+  `_pragma=busy_timeout(5000)` to any `db_url`, so a write that finds another
+  program writing to the file waits up to five seconds for it instead of
+  failing. WAL mode comes only from the URL: keep `_pragma=journal_mode(WAL)` in
+  one you set
 - `cache_max_bytes` — disk cache size limit in bytes; `0` disables the disk
   cache entirely; omitted defaults to 75% of the sum of the free space on the
   filesystem containing `<state_dir>/cache/` and the bytes of source and
diff --git a/TODO.md b/TODO.md
index a00b1ef..e8044db 100644
--- a/TODO.md
+++ b/TODO.md
@@ -62,6 +62,12 @@ P2: security: per-origin rate limiting
   another way, such as with `%2F` in the path, whatever else the path holds, or
   with a user name and password. It also answers 400 for a source path holding a
   `..` segment, as `/v1/image/` refuses one.
+- 2026-10-16 a `state_dir` whose name holds `%`, `?` or `#` keeps its database
+  at `state.sqlite3` in that directory (closes #290): the `db_url` derived from
+  it put the name into a `file:` URL as it was, so SQLite read `%41` as `A`, a
+  `?` or `#` ended the path early, and the journal mode parameter was lost. The
+  derived `db_url` is now built with Go's `net/url`, which writes those
+  characters as `%25`, `%3F` and `%23`.
 - 2026-10-15 a thin source requested with `fit=cover` no longer makes libvips
   build a huge image (closes #282): libvips' thumbnail scaled the whole source
   to cover the requested size and copied that into memory before it cut out the
diff --git a/internal/config/config.go b/internal/config/config.go
index ff4ac6c..d94fb74 100644
--- a/internal/config/config.go
+++ b/internal/config/config.go
@@ -338,8 +338,7 @@ func newFromSmartConfig(sc *smartconfig.Config) (*Config, error) {
 				settingName(keyDBURL), errValueEmpty)
 		}
 
-		// The driver sets the journal mode only through a _pragma parameter.
-		c.DBURL = fmt.Sprintf("file:%s/state.sqlite3?_pragma=journal_mode(WAL)", c.StateDir)
+		c.DBURL = defaultDBURL(c.StateDir)
 	}
 
 	if loader.err != nil {
@@ -354,6 +353,24 @@ func newFromSmartConfig(sc *smartconfig.Config) (*Config, error) {
 	return c, nil
 }
 
+// defaultDBURL returns the db_url used when db_url is omitted: the file
+// state.sqlite3 in stateDir, in WAL mode, which the driver sets only
+// through a _pragma parameter. String() percent-encodes any %, ? or # in
+// the directory name, which SQLite decodes back, so the URL names exactly
+// this file. OmitHost writes file:/var/lib/pixa/... instead of
+// file:///var/lib/pixa/..., and keeps a relative stateDir from being read
+// as a host name.
+func defaultDBURL(stateDir string) string {
+	dbURL := url.URL{
+		Scheme:   "file",
+		OmitHost: true,
+		Path:     stateDir + "/state.sqlite3",
+		RawQuery: "_pragma=journal_mode(WAL)",
+	}
+
+	return dbURL.String()
+}
+
 // validateKnownKeys rejects configuration files containing keys the
 // application does not understand, so typos fail at startup instead of
 // being silently ignored, and rejects keys that are explicitly set to