remove ENV overrides, shorten opt name, only allow setting via cmake option (not envvar), update doc ref

Author: tevio

cmake/script/CoverageInclude.cmake.in

@@ -2,13 +2,6 @@
 # Distributed under the MIT software license, see the accompanying
 # file COPYING or https://opensource.org/license/mit/.
 
-if(NOT DEFINED LCOV_OPTS)
-  set(LCOV_OPTS "$ENV{LCOV_OPTS}")
-endif()
-if(NOT DEFINED GENHTML_OPTS)
-  set(GENHTML_OPTS "$ENV{GENHTML_OPTS}")
-endif()
-
 if("@CMAKE_CXX_COMPILER_ID@" STREQUAL "Clang")
   find_program(LLVM_COV_EXECUTABLE llvm-cov REQUIRED)
   set(COV_TOOL "${LLVM_COV_EXECUTABLE} gcov")
@@ -31,8 +24,8 @@ separate_arguments(LCOV_OPTS)
 set(LCOV_COMMAND ${LCOV_EXECUTABLE} --gcov-tool ${CMAKE_CURRENT_LIST_DIR}/cov_tool_wrapper.sh ${LCOV_OPTS})
 
 find_program(GENHTML_EXECUTABLE genhtml REQUIRED)
-separate_arguments(GENHTML_OPTS)
-set(GENHTML_COMMAND ${GENHTML_EXECUTABLE} --show-details ${GENHTML_OPTS})
+separate_arguments(HTML_OPTS)
+set(GENHTML_COMMAND ${GENHTML_EXECUTABLE} --show-details ${HTML_OPTS})
 
 find_program(GREP_EXECUTABLE grep REQUIRED)
 find_program(AWK_EXECUTABLE awk REQUIRED)

doc/developer-notes.md

@@ -509,7 +509,11 @@ cmake -DLCOV_OPTS="--rc branch_coverage=1" -P build/Coverage.cmake
 ```
 
 GENHTML_OPTS can be specified to provide options to genhtml, the program that generates the
-html report from lcov: `GENHTML_OPTS="--exclude boost"`.
+html report from lcov: `HTML_OPTS="--exclude boost"`.
+
+```
+cmake -DHTML_OPTS="--exclude boost" -P build/Coverage.cmake
+```
 
 To enable test parallelism:
 ```

doc/leveldb.md

@@ -0,0 +1,38 @@
+LevelDB overview
+=================
+
+LevelDB is a fast, embeddable key–value storage library created at Google. It provides an ordered mapping from arbitrary byte-string keys to byte-string values, optimized for high write throughput and efficient range queries.
+
+Why Bitcoin Knots uses LevelDB
+------------------------------
+Bitcoin Knots (like Bitcoin Core) embeds LevelDB to persist several on-disk databases, for example:
+- The chainstate (UTXO set) and block index metadata
+- Various indexes and caches used by optional components
+
+LevelDB’s log-structured merge-tree (LSM) design offers:
+- High write performance with sequential disk access patterns
+- Background compaction for space and read amplification control
+- Forward/backward iteration and efficient prefix/range scans
+
+Key characteristics
+-------------------
+- API: Put(key, value), Get(key), Delete(key); batched atomic writes via WriteBatch; consistent point-in-time Snapshots; forward/backward Iterators.
+- Ordering: Keys are stored in sorted order (default bytewise comparator; custom comparators supported).
+- Compression: Supports Snappy compression when available.
+- Embedding: Single-process library; there is no built-in client/server mode.
+
+Limitations to be aware of
+--------------------------
+- Not a SQL or relational database; no secondary indexes or query language.
+- Single-process access to a given database path at a time.
+- Behavior and performance can be sensitive to filesystem and OS parameters (e.g., open file limits).
+
+Project references and further reading
+--------------------------------------
+- Upstream LevelDB documentation: src/leveldb/README.md (bundled) and https://github.com/google/leveldb/blob/master/doc/index.md
+- Bitcoin developer notes on "Upgrading LevelDB": doc/developer-notes.md#upgrading-leveldb — important considerations for compatibility, file descriptor usage, and consensus safety when changing LevelDB versions.
+
+Notes for contributors
+----------------------
+- Any upgrades to the embedded LevelDB subtree must follow the guidance in the developer notes to avoid consensus-affecting changes and resource limit issues.
+- CRC32C acceleration used by LevelDB is provided via src/crc32c; see developer notes for details.