test: add scenario for checking chunk-size

5 files changed, 53 insertions, 17 deletions

diff --git a/obnam.md b/obnam.md
index b41a118..34535a5 100644
--- a/obnam.md
+++ b/obnam.md
@@ -771,6 +771,10 @@ The server has the following API for managing chunks:
 * `GET /chunks?sha256=xyzzy` – find chunks on the server whose
   metadata indicates their contents has a given SHA256 checksum
 * `GET /chunks?generation=true` – find generation chunks
+* `GET /chunks?data=True` – find chunks with file data
+  - this is meant for testing only
+  - it excludes generation chunks, and chunks used to store the
+    generation's SQLite file
 
 HTTP status codes are used to indicate if a request succeeded or not,
 using the customary meanings.
@@ -858,6 +862,7 @@ when I POST data.dat to /chunks, with chunk-meta: {"sha256":"abc"}
 then HTTP status code is 201
 and content-type is application/json
 and the JSON body has a field chunk_id, henceforth ID
+and server has 1 file chunks
 ~~~
 
 We must be able to retrieve it.
@@ -1145,6 +1150,28 @@ given a manifest of the directory live restored in rest in rest.yaml
 then files live.yaml and rest.yaml match
 ~~~
 
+## Set chunk size
+
+This scenario verifies that the user can set the chunk size in the
+configuration file. The chunk size only affects the chunks of live
+data.
+
+~~~scenario
+given an installed obnam
+given a running chunk server
+given a client config based on tiny-chunk-size.yaml
+given a file live/data.dat containing "abc"
+when I run obnam --config tiny-chunk-size.yaml backup
+then server has 3 file chunks
+~~~
+
+~~~{#tiny-chunk-size.yaml .file .yaml .numberLines}
+verify_tls_cert: false
+root: live
+chunk_size: 1
+~~~
+
+
 ## Backup or not for the right reason
 
 The decision of whether to back up a file or keep the version in the
diff --git a/subplot/data.py b/subplot/data.py
index 2a54037..f3faf2b 100644
--- a/subplot/data.py
+++ b/subplot/data.py
@@ -5,14 +5,17 @@ import random
 import yaml
 
 
-def create_file_with_random_data(ctx, filename=None):
-    N = 128
-    data = "".join(chr(random.randint(0, 255)) for i in range(N)).encode("UTF-8")
+def create_file_with_given_data(ctx, filename=None, data=None):
+    logging.debug(f"creating file {filename} with {data!r}")
     dirname = os.path.dirname(filename) or "."
-    logging.debug(f"create_file_with_random_data: dirname={dirname}")
     os.makedirs(dirname, exist_ok=True)
-    with open(filename, "wb") as f:
-        f.write(data)
+    open(filename, "wb").write(data.encode("UTF-8"))
+
+
+def create_file_with_random_data(ctx, filename=None):
+    N = 128
+    data = "".join(chr(random.randint(0, 255)) for i in range(N))
+    create_file_with_given_data(ctx, filename=filename, data=data)
 
 
 def create_nonutf8_filename(ctx, dirname=None):
diff --git a/subplot/data.yaml b/subplot/data.yaml
index 6d384b8..9538daa 100644
--- a/subplot/data.yaml
+++ b/subplot/data.yaml
@@ -1,6 +1,4 @@
-- given: >
-    a file (?P<filename>\\S+) containing "(?P<data>.*)"
-  regex: true
+- given: a file {filename} containing "{data:text}"
   function: create_file_with_given_data
 
 - given: "a file {filename} containing some random data"
diff --git a/subplot/server.py b/subplot/server.py
index 289e181..df594f7 100644
--- a/subplot/server.py
+++ b/subplot/server.py
@@ -5,8 +5,6 @@ import random
 import re
 import requests
 import shutil
-import socket
-import time
 import urllib3
 import yaml
 
@@ -35,7 +33,9 @@ def start_chunk_server(ctx):
         "address": f"localhost:{port}",
     }
 
-    server_binary = os.path.abspath(os.path.join(srcdir, "target", "debug", "obnam-server"))
+    server_binary = os.path.abspath(
+        os.path.join(srcdir, "target", "debug", "obnam-server")
+    )
 
     filename = "config.yaml"
     yaml.safe_dump(config, stream=open(filename, "w"))
@@ -44,11 +44,7 @@ def start_chunk_server(ctx):
     ctx["server_url"] = f"https://{config['address']}"
 
     daemon_start_on_port(
-        ctx,
-        name="obnam-server",
-        path=server_binary,
-        args=filename,
-        port=port,
+        ctx, name="obnam-server", path=server_binary, args=filename, port=port
     )
 
 
@@ -138,6 +134,15 @@ def json_body_matches(ctx, wanted=None):
         assert_eq(body.get(key, "not.there"), wanted[key])
 
 
+def server_has_n_file_chunks(ctx, n=None):
+    assert_eq = globals()["assert_eq"]
+    n = int(n)
+    url = f"{ctx['server_url']}/chunks?data=true"
+    _request(ctx, requests.get, url)
+    num_chunks = len(ctx["http.json"])
+    assert_eq(n, num_chunks)
+
+
 # Make an HTTP request.
 def _request(ctx, method, url, headers=None, data=None):
     r = method(url, headers=headers, data=data, verify=False)
diff --git a/subplot/server.yaml b/subplot/server.yaml
index 68f8f0c..60f8a44 100644
--- a/subplot/server.yaml
+++ b/subplot/server.yaml
@@ -43,3 +43,6 @@
 
 - then: "the body matches file {filename}"
   function: body_matches_file
+
+- then: "server has {n:int} file chunks"
+  function: server_has_n_file_chunks